diff --git a/src/hotspot/cpu/aarch64/c1_CodeStubs_aarch64.cpp b/src/hotspot/cpu/aarch64/c1_CodeStubs_aarch64.cpp
index ca175fe1c47b1..89a97a4984fc8 100644
--- a/src/hotspot/cpu/aarch64/c1_CodeStubs_aarch64.cpp
+++ b/src/hotspot/cpu/aarch64/c1_CodeStubs_aarch64.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999, 2023, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 1999, 2024, Oracle and/or its affiliates. All rights reserved.
  * Copyright (c) 2014, Red Hat Inc. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
@@ -57,7 +57,7 @@ void CounterOverflowStub::emit_code(LIR_Assembler* ce) {
   __ mov_metadata(rscratch1, m);
   ce->store_parameter(rscratch1, 1);
   ce->store_parameter(_bci, 0);
-  __ far_call(RuntimeAddress(Runtime1::entry_for(Runtime1::counter_overflow_id)));
+  __ far_call(RuntimeAddress(Runtime1::entry_for(C1StubId::counter_overflow_id)));
   ce->add_call_info_here(_info);
   ce->verify_oop_map(_info);
   __ b(_continuation);
@@ -66,7 +66,7 @@ void CounterOverflowStub::emit_code(LIR_Assembler* ce) {
 void RangeCheckStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
   if (_info->deoptimize_on_exception()) {
-    address a = Runtime1::entry_for(Runtime1::predicate_failed_trap_id);
+    address a = Runtime1::entry_for(C1StubId::predicate_failed_trap_id);
     __ far_call(RuntimeAddress(a));
     ce->add_call_info_here(_info);
     ce->verify_oop_map(_info);
@@ -79,13 +79,13 @@ void RangeCheckStub::emit_code(LIR_Assembler* ce) {
   } else {
     __ mov(rscratch1, _index->as_jint());
   }
-  Runtime1::StubID stub_id;
+  C1StubId stub_id;
   if (_throw_index_out_of_bounds_exception) {
-    stub_id = Runtime1::throw_index_exception_id;
+    stub_id = C1StubId::throw_index_exception_id;
   } else {
     assert(_array != LIR_Opr::nullOpr(), "sanity");
     __ mov(rscratch2, _array->as_pointer_register());
-    stub_id = Runtime1::throw_range_check_failed_id;
+    stub_id = C1StubId::throw_range_check_failed_id;
   }
   __ lea(lr, RuntimeAddress(Runtime1::entry_for(stub_id)));
   __ blr(lr);
@@ -100,7 +100,7 @@ PredicateFailedStub::PredicateFailedStub(CodeEmitInfo* info) {
 
 void PredicateFailedStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
-  address a = Runtime1::entry_for(Runtime1::predicate_failed_trap_id);
+  address a = Runtime1::entry_for(C1StubId::predicate_failed_trap_id);
   __ far_call(RuntimeAddress(a));
   ce->add_call_info_here(_info);
   ce->verify_oop_map(_info);
@@ -112,7 +112,7 @@ void DivByZeroStub::emit_code(LIR_Assembler* ce) {
     ce->compilation()->implicit_exception_table()->append(_offset, __ offset());
   }
   __ bind(_entry);
-  __ far_call(RuntimeAddress(Runtime1::entry_for(Runtime1::throw_div0_exception_id)));
+  __ far_call(RuntimeAddress(Runtime1::entry_for(C1StubId::throw_div0_exception_id)));
   ce->add_call_info_here(_info);
   ce->verify_oop_map(_info);
 #ifdef ASSERT
@@ -124,14 +124,14 @@ void DivByZeroStub::emit_code(LIR_Assembler* ce) {
 
 // Implementation of NewInstanceStub
 
-NewInstanceStub::NewInstanceStub(LIR_Opr klass_reg, LIR_Opr result, ciInstanceKlass* klass, CodeEmitInfo* info, Runtime1::StubID stub_id) {
+NewInstanceStub::NewInstanceStub(LIR_Opr klass_reg, LIR_Opr result, ciInstanceKlass* klass, CodeEmitInfo* info, C1StubId stub_id) {
   _result = result;
   _klass = klass;
   _klass_reg = klass_reg;
   _info = new CodeEmitInfo(info);
-  assert(stub_id == Runtime1::new_instance_id                 ||
-         stub_id == Runtime1::fast_new_instance_id            ||
-         stub_id == Runtime1::fast_new_instance_init_check_id,
+  assert(stub_id == C1StubId::new_instance_id                 ||
+         stub_id == C1StubId::fast_new_instance_id            ||
+         stub_id == C1StubId::fast_new_instance_init_check_id,
          "need new_instance id");
   _stub_id   = stub_id;
 }
@@ -167,7 +167,7 @@ void NewTypeArrayStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
   assert(_length->as_register() == r19, "length must in r19,");
   assert(_klass_reg->as_register() == r3, "klass_reg must in r3");
-  __ far_call(RuntimeAddress(Runtime1::entry_for(Runtime1::new_type_array_id)));
+  __ far_call(RuntimeAddress(Runtime1::entry_for(C1StubId::new_type_array_id)));
   ce->add_call_info_here(_info);
   ce->verify_oop_map(_info);
   assert(_result->as_register() == r0, "result must in r0");
@@ -190,7 +190,7 @@ void NewObjectArrayStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
   assert(_length->as_register() == r19, "length must in r19,");
   assert(_klass_reg->as_register() == r3, "klass_reg must in r3");
-  __ far_call(RuntimeAddress(Runtime1::entry_for(Runtime1::new_object_array_id)));
+  __ far_call(RuntimeAddress(Runtime1::entry_for(C1StubId::new_object_array_id)));
   ce->add_call_info_here(_info);
   ce->verify_oop_map(_info);
   assert(_result->as_register() == r0, "result must in r0");
@@ -202,11 +202,11 @@ void MonitorEnterStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
   ce->store_parameter(_obj_reg->as_register(),  1);
   ce->store_parameter(_lock_reg->as_register(), 0);
-  Runtime1::StubID enter_id;
+  C1StubId enter_id;
   if (ce->compilation()->has_fpu_code()) {
-    enter_id = Runtime1::monitorenter_id;
+    enter_id = C1StubId::monitorenter_id;
   } else {
-    enter_id = Runtime1::monitorenter_nofpu_id;
+    enter_id = C1StubId::monitorenter_nofpu_id;
   }
   __ far_call(RuntimeAddress(Runtime1::entry_for(enter_id)));
   ce->add_call_info_here(_info);
@@ -223,11 +223,11 @@ void MonitorExitStub::emit_code(LIR_Assembler* ce) {
   }
   ce->store_parameter(_lock_reg->as_register(), 0);
   // note: non-blocking leaf routine => no call info needed
-  Runtime1::StubID exit_id;
+  C1StubId exit_id;
   if (ce->compilation()->has_fpu_code()) {
-    exit_id = Runtime1::monitorexit_id;
+    exit_id = C1StubId::monitorexit_id;
   } else {
-    exit_id = Runtime1::monitorexit_nofpu_id;
+    exit_id = C1StubId::monitorexit_nofpu_id;
   }
   __ adr(lr, _continuation);
   __ far_jump(RuntimeAddress(Runtime1::entry_for(exit_id)));
@@ -255,7 +255,7 @@ void PatchingStub::emit_code(LIR_Assembler* ce) {
 void DeoptimizeStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
   ce->store_parameter(_trap_request, 0);
-  __ far_call(RuntimeAddress(Runtime1::entry_for(Runtime1::deoptimize_id)));
+  __ far_call(RuntimeAddress(Runtime1::entry_for(C1StubId::deoptimize_id)));
   ce->add_call_info_here(_info);
   DEBUG_ONLY(__ should_not_reach_here());
 }
@@ -265,9 +265,9 @@ void ImplicitNullCheckStub::emit_code(LIR_Assembler* ce) {
   address a;
   if (_info->deoptimize_on_exception()) {
     // Deoptimize, do not throw the exception, because it is probably wrong to do it here.
-    a = Runtime1::entry_for(Runtime1::predicate_failed_trap_id);
+    a = Runtime1::entry_for(C1StubId::predicate_failed_trap_id);
   } else {
-    a = Runtime1::entry_for(Runtime1::throw_null_pointer_exception_id);
+    a = Runtime1::entry_for(C1StubId::throw_null_pointer_exception_id);
   }
 
   ce->compilation()->implicit_exception_table()->append(_offset, __ offset());
diff --git a/src/hotspot/cpu/aarch64/c1_LIRAssembler_aarch64.cpp b/src/hotspot/cpu/aarch64/c1_LIRAssembler_aarch64.cpp
index 91430be5835b5..5e116d82761ac 100644
--- a/src/hotspot/cpu/aarch64/c1_LIRAssembler_aarch64.cpp
+++ b/src/hotspot/cpu/aarch64/c1_LIRAssembler_aarch64.cpp
@@ -321,19 +321,19 @@ void LIR_Assembler::deoptimize_trap(CodeEmitInfo *info) {
 
   switch (patching_id(info)) {
   case PatchingStub::access_field_id:
-    target = Runtime1::entry_for(Runtime1::access_field_patching_id);
+    target = Runtime1::entry_for(C1StubId::access_field_patching_id);
     reloc_type = relocInfo::section_word_type;
     break;
   case PatchingStub::load_klass_id:
-    target = Runtime1::entry_for(Runtime1::load_klass_patching_id);
+    target = Runtime1::entry_for(C1StubId::load_klass_patching_id);
     reloc_type = relocInfo::metadata_type;
     break;
   case PatchingStub::load_mirror_id:
-    target = Runtime1::entry_for(Runtime1::load_mirror_patching_id);
+    target = Runtime1::entry_for(C1StubId::load_mirror_patching_id);
     reloc_type = relocInfo::oop_type;
     break;
   case PatchingStub::load_appendix_id:
-    target = Runtime1::entry_for(Runtime1::load_appendix_patching_id);
+    target = Runtime1::entry_for(C1StubId::load_appendix_patching_id);
     reloc_type = relocInfo::oop_type;
     break;
   default: ShouldNotReachHere();
@@ -375,7 +375,7 @@ int LIR_Assembler::emit_exception_handler() {
   __ verify_not_null_oop(r0);
 
   // search an exception handler (r0: exception oop, r3: throwing pc)
-  __ far_call(RuntimeAddress(Runtime1::entry_for(Runtime1::handle_exception_from_callee_id)));
+  __ far_call(RuntimeAddress(Runtime1::entry_for(C1StubId::handle_exception_from_callee_id)));
   __ should_not_reach_here();
   guarantee(code_offset() - offset <= exception_handler_size(), "overflow");
   __ end_a_stub();
@@ -432,7 +432,7 @@ int LIR_Assembler::emit_unwind_handler() {
   // remove the activation and dispatch to the unwind handler
   __ block_comment("remove_frame and dispatch to the unwind handler");
   __ remove_frame(initial_frame_size_in_bytes());
-  __ far_jump(RuntimeAddress(Runtime1::entry_for(Runtime1::unwind_exception_id)));
+  __ far_jump(RuntimeAddress(Runtime1::entry_for(C1StubId::unwind_exception_id)));
 
   // Emit the slow path assembly
   if (stub != nullptr) {
@@ -875,19 +875,19 @@ void LIR_Assembler::klass2reg_with_patching(Register reg, CodeEmitInfo* info) {
 
   switch (patching_id(info)) {
   case PatchingStub::access_field_id:
-    target = Runtime1::entry_for(Runtime1::access_field_patching_id);
+    target = Runtime1::entry_for(C1StubId::access_field_patching_id);
     reloc_type = relocInfo::section_word_type;
     break;
   case PatchingStub::load_klass_id:
-    target = Runtime1::entry_for(Runtime1::load_klass_patching_id);
+    target = Runtime1::entry_for(C1StubId::load_klass_patching_id);
     reloc_type = relocInfo::metadata_type;
     break;
   case PatchingStub::load_mirror_id:
-    target = Runtime1::entry_for(Runtime1::load_mirror_patching_id);
+    target = Runtime1::entry_for(C1StubId::load_mirror_patching_id);
     reloc_type = relocInfo::oop_type;
     break;
   case PatchingStub::load_appendix_id:
-    target = Runtime1::entry_for(Runtime1::load_appendix_patching_id);
+    target = Runtime1::entry_for(C1StubId::load_appendix_patching_id);
     reloc_type = relocInfo::oop_type;
     break;
   default: ShouldNotReachHere();
@@ -1356,7 +1356,7 @@ void LIR_Assembler::emit_typecheck_helper(LIR_OpTypeCheck *op, Label* success, L
         __ br(Assembler::EQ, *success_target);
 
         __ stp(klass_RInfo, k_RInfo, Address(__ pre(sp, -2 * wordSize)));
-        __ far_call(RuntimeAddress(Runtime1::entry_for(Runtime1::slow_subtype_check_id)));
+        __ far_call(RuntimeAddress(Runtime1::entry_for(C1StubId::slow_subtype_check_id)));
         __ ldr(klass_RInfo, Address(__ post(sp, 2 * wordSize)));
         // result is a boolean
         __ cbzw(klass_RInfo, *failure_target);
@@ -1367,7 +1367,7 @@ void LIR_Assembler::emit_typecheck_helper(LIR_OpTypeCheck *op, Label* success, L
       __ check_klass_subtype_fast_path(klass_RInfo, k_RInfo, Rtmp1, success_target, failure_target, nullptr);
       // call out-of-line instance of __ check_klass_subtype_slow_path(...):
       __ stp(klass_RInfo, k_RInfo, Address(__ pre(sp, -2 * wordSize)));
-      __ far_call(RuntimeAddress(Runtime1::entry_for(Runtime1::slow_subtype_check_id)));
+      __ far_call(RuntimeAddress(Runtime1::entry_for(C1StubId::slow_subtype_check_id)));
       __ ldp(k_RInfo, klass_RInfo, Address(__ post(sp, 2 * wordSize)));
       // result is a boolean
       __ cbz(k_RInfo, *failure_target);
@@ -1446,7 +1446,7 @@ void LIR_Assembler::emit_opTypeCheck(LIR_OpTypeCheck* op) {
     __ check_klass_subtype_fast_path(klass_RInfo, k_RInfo, Rtmp1, success_target, failure_target, nullptr);
     // call out-of-line instance of __ check_klass_subtype_slow_path(...):
     __ stp(klass_RInfo, k_RInfo, Address(__ pre(sp, -2 * wordSize)));
-    __ far_call(RuntimeAddress(Runtime1::entry_for(Runtime1::slow_subtype_check_id)));
+    __ far_call(RuntimeAddress(Runtime1::entry_for(C1StubId::slow_subtype_check_id)));
     __ ldp(k_RInfo, klass_RInfo, Address(__ post(sp, 2 * wordSize)));
     // result is a boolean
     __ cbzw(k_RInfo, *failure_target);
@@ -2035,7 +2035,7 @@ void LIR_Assembler::throw_op(LIR_Opr exceptionPC, LIR_Opr exceptionOop, CodeEmit
   // exception object is not added to oop map by LinearScan
   // (LinearScan assumes that no oops are in fixed registers)
   info->add_register_oop(exceptionOop);
-  Runtime1::StubID unwind_id;
+  C1StubId unwind_id;
 
   // get current pc information
   // pc is only needed if the method has an exception handler, the unwind code does not need it.
@@ -2054,9 +2054,9 @@ void LIR_Assembler::throw_op(LIR_Opr exceptionPC, LIR_Opr exceptionOop, CodeEmit
   __ verify_not_null_oop(r0);
   // search an exception handler (r0: exception oop, r3: throwing pc)
   if (compilation()->has_fpu_code()) {
-    unwind_id = Runtime1::handle_exception_id;
+    unwind_id = C1StubId::handle_exception_id;
   } else {
-    unwind_id = Runtime1::handle_exception_nofpu_id;
+    unwind_id = C1StubId::handle_exception_nofpu_id;
   }
   __ far_call(RuntimeAddress(Runtime1::entry_for(unwind_id)));
 
@@ -2337,7 +2337,7 @@ void LIR_Assembler::emit_arraycopy(LIR_OpArrayCopy* op) {
       __ check_klass_subtype_fast_path(src, dst, tmp, &cont, &slow, nullptr);
 
       __ PUSH(src, dst);
-      __ far_call(RuntimeAddress(Runtime1::entry_for(Runtime1::slow_subtype_check_id)));
+      __ far_call(RuntimeAddress(Runtime1::entry_for(C1StubId::slow_subtype_check_id)));
       __ POP(src, dst);
 
       __ cbnz(src, cont);
diff --git a/src/hotspot/cpu/aarch64/c1_LIRGenerator_aarch64.cpp b/src/hotspot/cpu/aarch64/c1_LIRGenerator_aarch64.cpp
index 8f1260feba3ea..4acac65ad5bab 100644
--- a/src/hotspot/cpu/aarch64/c1_LIRGenerator_aarch64.cpp
+++ b/src/hotspot/cpu/aarch64/c1_LIRGenerator_aarch64.cpp
@@ -1246,7 +1246,7 @@ void LIRGenerator::do_NewMultiArray(NewMultiArray* x) {
   args->append(rank);
   args->append(varargs);
   LIR_Opr reg = result_register_for(x->type());
-  __ call_runtime(Runtime1::entry_for(Runtime1::new_multi_array_id),
+  __ call_runtime(Runtime1::entry_for(C1StubId::new_multi_array_id),
                   LIR_OprFact::illegalOpr,
                   reg, args, info);
 
@@ -1277,14 +1277,14 @@ void LIRGenerator::do_CheckCast(CheckCast* x) {
   CodeStub* stub;
   if (x->is_incompatible_class_change_check()) {
     assert(patching_info == nullptr, "can't patch this");
-    stub = new SimpleExceptionStub(Runtime1::throw_incompatible_class_change_error_id, LIR_OprFact::illegalOpr, info_for_exception);
+    stub = new SimpleExceptionStub(C1StubId::throw_incompatible_class_change_error_id, LIR_OprFact::illegalOpr, info_for_exception);
   } else if (x->is_invokespecial_receiver_check()) {
     assert(patching_info == nullptr, "can't patch this");
     stub = new DeoptimizeStub(info_for_exception,
                               Deoptimization::Reason_class_check,
                               Deoptimization::Action_none);
   } else {
-    stub = new SimpleExceptionStub(Runtime1::throw_class_cast_exception_id, obj.result(), info_for_exception);
+    stub = new SimpleExceptionStub(C1StubId::throw_class_cast_exception_id, obj.result(), info_for_exception);
   }
   LIR_Opr reg = rlock_result(x);
   LIR_Opr tmp3 = LIR_OprFact::illegalOpr;
diff --git a/src/hotspot/cpu/aarch64/c1_MacroAssembler_aarch64.cpp b/src/hotspot/cpu/aarch64/c1_MacroAssembler_aarch64.cpp
index d0d11d437e83e..8d1b3902ce42e 100644
--- a/src/hotspot/cpu/aarch64/c1_MacroAssembler_aarch64.cpp
+++ b/src/hotspot/cpu/aarch64/c1_MacroAssembler_aarch64.cpp
@@ -267,7 +267,7 @@ void C1_MacroAssembler::initialize_object(Register obj, Register klass, Register
 
   if (CURRENT_ENV->dtrace_alloc_probes()) {
     assert(obj == r0, "must be");
-    far_call(RuntimeAddress(Runtime1::entry_for(Runtime1::dtrace_object_alloc_id)));
+    far_call(RuntimeAddress(Runtime1::entry_for(C1StubId::dtrace_object_alloc_id)));
   }
 
   verify_oop(obj);
@@ -308,7 +308,7 @@ void C1_MacroAssembler::allocate_array(Register obj, Register len, Register t1,
 
   if (CURRENT_ENV->dtrace_alloc_probes()) {
     assert(obj == r0, "must be");
-    far_call(RuntimeAddress(Runtime1::entry_for(Runtime1::dtrace_object_alloc_id)));
+    far_call(RuntimeAddress(Runtime1::entry_for(C1StubId::dtrace_object_alloc_id)));
   }
 
   verify_oop(obj);
diff --git a/src/hotspot/cpu/aarch64/c1_Runtime1_aarch64.cpp b/src/hotspot/cpu/aarch64/c1_Runtime1_aarch64.cpp
index cb9eb03c580d2..0b9acc0f3a885 100644
--- a/src/hotspot/cpu/aarch64/c1_Runtime1_aarch64.cpp
+++ b/src/hotspot/cpu/aarch64/c1_Runtime1_aarch64.cpp
@@ -100,10 +100,10 @@ int StubAssembler::call_RT(Register oop_result1, Register metadata_result, addre
     if (frame_size() == no_frame_size) {
       leave();
       far_jump(RuntimeAddress(StubRoutines::forward_exception_entry()));
-    } else if (_stub_id == Runtime1::forward_exception_id) {
+    } else if (_stub_id == (int)C1StubId::forward_exception_id) {
       should_not_reach_here();
     } else {
-      far_jump(RuntimeAddress(Runtime1::entry_for(Runtime1::forward_exception_id)));
+      far_jump(RuntimeAddress(Runtime1::entry_for(C1StubId::forward_exception_id)));
     }
     bind(L);
   }
@@ -358,7 +358,7 @@ OopMapSet* Runtime1::generate_exception_throw(StubAssembler* sasm, address targe
 }
 
 
-OopMapSet* Runtime1::generate_handle_exception(StubID id, StubAssembler *sasm) {
+OopMapSet* Runtime1::generate_handle_exception(C1StubId id, StubAssembler *sasm) {
   __ block_comment("generate_handle_exception");
 
   // incoming parameters
@@ -370,7 +370,7 @@ OopMapSet* Runtime1::generate_handle_exception(StubID id, StubAssembler *sasm) {
   OopMapSet* oop_maps = new OopMapSet();
   OopMap* oop_map = nullptr;
   switch (id) {
-  case forward_exception_id:
+  case C1StubId::forward_exception_id:
     // We're handling an exception in the context of a compiled frame.
     // The registers have been saved in the standard places.  Perform
     // an exception lookup in the caller and dispatch to the handler
@@ -390,12 +390,12 @@ OopMapSet* Runtime1::generate_handle_exception(StubID id, StubAssembler *sasm) {
     __ str(zr, Address(rthread, JavaThread::vm_result_offset()));
     __ str(zr, Address(rthread, JavaThread::vm_result_2_offset()));
     break;
-  case handle_exception_nofpu_id:
-  case handle_exception_id:
+  case C1StubId::handle_exception_nofpu_id:
+  case C1StubId::handle_exception_id:
     // At this point all registers MAY be live.
-    oop_map = save_live_registers(sasm, id != handle_exception_nofpu_id);
+    oop_map = save_live_registers(sasm, id != C1StubId::handle_exception_nofpu_id);
     break;
-  case handle_exception_from_callee_id: {
+  case C1StubId::handle_exception_from_callee_id: {
     // At this point all registers except exception oop (r0) and
     // exception pc (lr) are dead.
     const int frame_size = 2 /*fp, return address*/;
@@ -453,13 +453,13 @@ OopMapSet* Runtime1::generate_handle_exception(StubID id, StubAssembler *sasm) {
   __ str(r0, Address(rfp, 1*BytesPerWord));
 
   switch (id) {
-  case forward_exception_id:
-  case handle_exception_nofpu_id:
-  case handle_exception_id:
+  case C1StubId::forward_exception_id:
+  case C1StubId::handle_exception_nofpu_id:
+  case C1StubId::handle_exception_id:
     // Restore the registers that were saved at the beginning.
-    restore_live_registers(sasm, id != handle_exception_nofpu_id);
+    restore_live_registers(sasm, id != C1StubId::handle_exception_nofpu_id);
     break;
-  case handle_exception_from_callee_id:
+  case C1StubId::handle_exception_from_callee_id:
     break;
   default:  ShouldNotReachHere();
   }
@@ -611,7 +611,7 @@ OopMapSet* Runtime1::generate_patching(StubAssembler* sasm, address target) {
 }
 
 
-OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
+OopMapSet* Runtime1::generate_code_for(C1StubId id, StubAssembler* sasm) {
 
   const Register exception_oop = r0;
   const Register exception_pc  = r3;
@@ -628,7 +628,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
   OopMap* oop_map = nullptr;
   switch (id) {
     {
-    case forward_exception_id:
+    case C1StubId::forward_exception_id:
       {
         oop_maps = generate_handle_exception(id, sasm);
         __ leave();
@@ -636,31 +636,31 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case throw_div0_exception_id:
+    case C1StubId::throw_div0_exception_id:
       { StubFrame f(sasm, "throw_div0_exception", dont_gc_arguments, does_not_return);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_div0_exception), false);
       }
       break;
 
-    case throw_null_pointer_exception_id:
+    case C1StubId::throw_null_pointer_exception_id:
       { StubFrame f(sasm, "throw_null_pointer_exception", dont_gc_arguments, does_not_return);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_null_pointer_exception), false);
       }
       break;
 
-    case new_instance_id:
-    case fast_new_instance_id:
-    case fast_new_instance_init_check_id:
+    case C1StubId::new_instance_id:
+    case C1StubId::fast_new_instance_id:
+    case C1StubId::fast_new_instance_init_check_id:
       {
         Register klass = r3; // Incoming
         Register obj   = r0; // Result
 
-        if (id == new_instance_id) {
+        if (id == C1StubId::new_instance_id) {
           __ set_info("new_instance", dont_gc_arguments);
-        } else if (id == fast_new_instance_id) {
+        } else if (id == C1StubId::fast_new_instance_id) {
           __ set_info("fast new_instance", dont_gc_arguments);
         } else {
-          assert(id == fast_new_instance_init_check_id, "bad StubID");
+          assert(id == C1StubId::fast_new_instance_init_check_id, "bad C1StubId");
           __ set_info("fast new_instance init check", dont_gc_arguments);
         }
 
@@ -679,7 +679,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
 
       break;
 
-    case counter_overflow_id:
+    case C1StubId::counter_overflow_id:
       {
         Register bci = r0, method = r1;
         __ enter();
@@ -697,14 +697,14 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case new_type_array_id:
-    case new_object_array_id:
+    case C1StubId::new_type_array_id:
+    case C1StubId::new_object_array_id:
       {
         Register length   = r19; // Incoming
         Register klass    = r3; // Incoming
         Register obj      = r0; // Result
 
-        if (id == new_type_array_id) {
+        if (id == C1StubId::new_type_array_id) {
           __ set_info("new_type_array", dont_gc_arguments);
         } else {
           __ set_info("new_object_array", dont_gc_arguments);
@@ -717,7 +717,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
           Register t0 = obj;
           __ ldrw(t0, Address(klass, Klass::layout_helper_offset()));
           __ asrw(t0, t0, Klass::_lh_array_tag_shift);
-          int tag = ((id == new_type_array_id)
+          int tag = ((id == C1StubId::new_type_array_id)
                      ? Klass::_lh_array_tag_type_value
                      : Klass::_lh_array_tag_obj_value);
           __ mov(rscratch1, tag);
@@ -732,7 +732,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
         __ enter();
         OopMap* map = save_live_registers(sasm);
         int call_offset;
-        if (id == new_type_array_id) {
+        if (id == C1StubId::new_type_array_id) {
           call_offset = __ call_RT(obj, noreg, CAST_FROM_FN_PTR(address, new_type_array), klass, length);
         } else {
           call_offset = __ call_RT(obj, noreg, CAST_FROM_FN_PTR(address, new_object_array), klass, length);
@@ -750,7 +750,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case new_multi_array_id:
+    case C1StubId::new_multi_array_id:
       { StubFrame f(sasm, "new_multi_array", dont_gc_arguments);
         // r0,: klass
         // r19,: rank
@@ -770,7 +770,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case register_finalizer_id:
+    case C1StubId::register_finalizer_id:
       {
         __ set_info("register_finalizer", dont_gc_arguments);
 
@@ -802,19 +802,19 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case throw_class_cast_exception_id:
+    case C1StubId::throw_class_cast_exception_id:
       { StubFrame f(sasm, "throw_class_cast_exception", dont_gc_arguments, does_not_return);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_class_cast_exception), true);
       }
       break;
 
-    case throw_incompatible_class_change_error_id:
+    case C1StubId::throw_incompatible_class_change_error_id:
       { StubFrame f(sasm, "throw_incompatible_class_cast_exception", dont_gc_arguments, does_not_return);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_incompatible_class_change_error), false);
       }
       break;
 
-    case slow_subtype_check_id:
+    case C1StubId::slow_subtype_check_id:
       {
         // Typical calling sequence:
         // __ push(klass_RInfo);  // object klass or other subclass
@@ -857,10 +857,10 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case monitorenter_nofpu_id:
+    case C1StubId::monitorenter_nofpu_id:
       save_fpu_registers = false;
       // fall through
-    case monitorenter_id:
+    case C1StubId::monitorenter_id:
       {
         StubFrame f(sasm, "monitorenter", dont_gc_arguments);
         OopMap* map = save_live_registers(sasm, save_fpu_registers);
@@ -878,10 +878,10 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case monitorexit_nofpu_id:
+    case C1StubId::monitorexit_nofpu_id:
       save_fpu_registers = false;
       // fall through
-    case monitorexit_id:
+    case C1StubId::monitorexit_id:
       {
         StubFrame f(sasm, "monitorexit", dont_gc_arguments);
         OopMap* map = save_live_registers(sasm, save_fpu_registers);
@@ -901,7 +901,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case deoptimize_id:
+    case C1StubId::deoptimize_id:
       {
         StubFrame f(sasm, "deoptimize", dont_gc_arguments, does_not_return);
         OopMap* oop_map = save_live_registers(sasm);
@@ -918,13 +918,13 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case throw_range_check_failed_id:
+    case C1StubId::throw_range_check_failed_id:
       { StubFrame f(sasm, "range_check_failed", dont_gc_arguments, does_not_return);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_range_check_exception), true);
       }
       break;
 
-    case unwind_exception_id:
+    case C1StubId::unwind_exception_id:
       { __ set_info("unwind_exception", dont_gc_arguments);
         // note: no stubframe since we are about to leave the current
         //       activation and we are calling a leaf VM function only.
@@ -932,54 +932,54 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case access_field_patching_id:
+    case C1StubId::access_field_patching_id:
       { StubFrame f(sasm, "access_field_patching", dont_gc_arguments, does_not_return);
         // we should set up register map
         oop_maps = generate_patching(sasm, CAST_FROM_FN_PTR(address, access_field_patching));
       }
       break;
 
-    case load_klass_patching_id:
+    case C1StubId::load_klass_patching_id:
       { StubFrame f(sasm, "load_klass_patching", dont_gc_arguments, does_not_return);
         // we should set up register map
         oop_maps = generate_patching(sasm, CAST_FROM_FN_PTR(address, move_klass_patching));
       }
       break;
 
-    case load_mirror_patching_id:
+    case C1StubId::load_mirror_patching_id:
       { StubFrame f(sasm, "load_mirror_patching", dont_gc_arguments, does_not_return);
         // we should set up register map
         oop_maps = generate_patching(sasm, CAST_FROM_FN_PTR(address, move_mirror_patching));
       }
       break;
 
-    case load_appendix_patching_id:
+    case C1StubId::load_appendix_patching_id:
       { StubFrame f(sasm, "load_appendix_patching", dont_gc_arguments, does_not_return);
         // we should set up register map
         oop_maps = generate_patching(sasm, CAST_FROM_FN_PTR(address, move_appendix_patching));
       }
       break;
 
-    case handle_exception_nofpu_id:
-    case handle_exception_id:
+    case C1StubId::handle_exception_nofpu_id:
+    case C1StubId::handle_exception_id:
       { StubFrame f(sasm, "handle_exception", dont_gc_arguments);
         oop_maps = generate_handle_exception(id, sasm);
       }
       break;
 
-    case handle_exception_from_callee_id:
+    case C1StubId::handle_exception_from_callee_id:
       { StubFrame f(sasm, "handle_exception_from_callee", dont_gc_arguments);
         oop_maps = generate_handle_exception(id, sasm);
       }
       break;
 
-    case throw_index_exception_id:
+    case C1StubId::throw_index_exception_id:
       { StubFrame f(sasm, "index_range_check_failed", dont_gc_arguments, does_not_return);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_index_exception), true);
       }
       break;
 
-    case throw_array_store_exception_id:
+    case C1StubId::throw_array_store_exception_id:
       { StubFrame f(sasm, "throw_array_store_exception", dont_gc_arguments, does_not_return);
         // tos + 0: link
         //     + 1: return address
@@ -987,7 +987,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case predicate_failed_trap_id:
+    case C1StubId::predicate_failed_trap_id:
       {
         StubFrame f(sasm, "predicate_failed_trap", dont_gc_arguments, does_not_return);
 
@@ -1005,7 +1005,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case dtrace_object_alloc_id:
+    case C1StubId::dtrace_object_alloc_id:
       { // c_rarg0: object
         StubFrame f(sasm, "dtrace_object_alloc", dont_gc_arguments);
         save_live_registers(sasm);
diff --git a/src/hotspot/cpu/arm/c1_CodeStubs_arm.cpp b/src/hotspot/cpu/arm/c1_CodeStubs_arm.cpp
index 3d8dbc38071ed..8e85fa88a8749 100644
--- a/src/hotspot/cpu/arm/c1_CodeStubs_arm.cpp
+++ b/src/hotspot/cpu/arm/c1_CodeStubs_arm.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2008, 2023, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2008, 2024, Oracle and/or its affiliates. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -46,7 +46,7 @@ void CounterOverflowStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
   ce->store_parameter(_bci, 0);
   ce->store_parameter(_method->as_constant_ptr()->as_metadata(), 1);
-  __ call(Runtime1::entry_for(Runtime1::counter_overflow_id), relocInfo::runtime_call_type);
+  __ call(Runtime1::entry_for(C1StubId::counter_overflow_id), relocInfo::runtime_call_type);
   ce->add_call_info_here(_info);
   ce->verify_oop_map(_info);
 
@@ -57,7 +57,7 @@ void RangeCheckStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
 
   if (_info->deoptimize_on_exception()) {
-    __ call(Runtime1::entry_for(Runtime1::predicate_failed_trap_id), relocInfo::runtime_call_type);
+    __ call(Runtime1::entry_for(C1StubId::predicate_failed_trap_id), relocInfo::runtime_call_type);
     ce->add_call_info_here(_info);
     ce->verify_oop_map(_info);
     debug_only(__ should_not_reach_here());
@@ -73,10 +73,10 @@ void RangeCheckStub::emit_code(LIR_Assembler* ce) {
   }
 
   if (_throw_index_out_of_bounds_exception) {
-    __ call(Runtime1::entry_for(Runtime1::throw_index_exception_id), relocInfo::runtime_call_type);
+    __ call(Runtime1::entry_for(C1StubId::throw_index_exception_id), relocInfo::runtime_call_type);
   } else {
     __ str(_array->as_pointer_register(), Address(SP, BytesPerWord)); // ??? Correct offset? Correct instruction?
-    __ call(Runtime1::entry_for(Runtime1::throw_range_check_failed_id), relocInfo::runtime_call_type);
+    __ call(Runtime1::entry_for(C1StubId::throw_range_check_failed_id), relocInfo::runtime_call_type);
   }
   ce->add_call_info_here(_info);
   ce->verify_oop_map(_info);
@@ -89,7 +89,7 @@ PredicateFailedStub::PredicateFailedStub(CodeEmitInfo* info) {
 
 void PredicateFailedStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
-  __ call(Runtime1::entry_for(Runtime1::predicate_failed_trap_id), relocInfo::runtime_call_type);
+  __ call(Runtime1::entry_for(C1StubId::predicate_failed_trap_id), relocInfo::runtime_call_type);
   ce->add_call_info_here(_info);
   ce->verify_oop_map(_info);
   debug_only(__ should_not_reach_here());
@@ -100,7 +100,7 @@ void DivByZeroStub::emit_code(LIR_Assembler* ce) {
     ce->compilation()->implicit_exception_table()->append(_offset, __ offset());
   }
   __ bind(_entry);
-  __ call(Runtime1::entry_for(Runtime1::throw_div0_exception_id),
+  __ call(Runtime1::entry_for(C1StubId::throw_div0_exception_id),
           relocInfo::runtime_call_type);
   ce->add_call_info_here(_info);
   DEBUG_ONLY(STOP("DivByZero");)
@@ -109,14 +109,14 @@ void DivByZeroStub::emit_code(LIR_Assembler* ce) {
 
 // Implementation of NewInstanceStub
 
-NewInstanceStub::NewInstanceStub(LIR_Opr klass_reg, LIR_Opr result, ciInstanceKlass* klass, CodeEmitInfo* info, Runtime1::StubID stub_id) {
+NewInstanceStub::NewInstanceStub(LIR_Opr klass_reg, LIR_Opr result, ciInstanceKlass* klass, CodeEmitInfo* info, C1StubId stub_id) {
   _result = result;
   _klass = klass;
   _klass_reg = klass_reg;
   _info = new CodeEmitInfo(info);
-  assert(stub_id == Runtime1::new_instance_id                 ||
-         stub_id == Runtime1::fast_new_instance_id            ||
-         stub_id == Runtime1::fast_new_instance_init_check_id,
+  assert(stub_id == C1StubId::new_instance_id                 ||
+         stub_id == C1StubId::fast_new_instance_id            ||
+         stub_id == C1StubId::fast_new_instance_init_check_id,
          "need new_instance id");
   _stub_id   = stub_id;
 }
@@ -148,7 +148,7 @@ void NewTypeArrayStub::emit_code(LIR_Assembler* ce) {
   assert(_klass_reg->as_register() == R1, "runtime call setup");
   assert(_length->as_register() == R2, "runtime call setup");
   __ bind(_entry);
-  __ call(Runtime1::entry_for(Runtime1::new_type_array_id), relocInfo::runtime_call_type);
+  __ call(Runtime1::entry_for(C1StubId::new_type_array_id), relocInfo::runtime_call_type);
   ce->add_call_info_here(_info);
   ce->verify_oop_map(_info);
   __ b(_continuation);
@@ -170,7 +170,7 @@ void NewObjectArrayStub::emit_code(LIR_Assembler* ce) {
   assert(_klass_reg->as_register() == R1, "runtime call setup");
   assert(_length->as_register() == R2, "runtime call setup");
   __ bind(_entry);
-  __ call(Runtime1::entry_for(Runtime1::new_object_array_id), relocInfo::runtime_call_type);
+  __ call(Runtime1::entry_for(C1StubId::new_object_array_id), relocInfo::runtime_call_type);
   ce->add_call_info_here(_info);
   ce->verify_oop_map(_info);
   __ b(_continuation);
@@ -189,9 +189,9 @@ void MonitorEnterStub::emit_code(LIR_Assembler* ce) {
     __ str(lock_reg, Address(SP, BytesPerWord));
   }
 
-  Runtime1::StubID enter_id = ce->compilation()->has_fpu_code() ?
-                              Runtime1::monitorenter_id :
-                              Runtime1::monitorenter_nofpu_id;
+  C1StubId enter_id = ce->compilation()->has_fpu_code() ?
+                              C1StubId::monitorenter_id :
+                              C1StubId::monitorenter_nofpu_id;
   __ call(Runtime1::entry_for(enter_id), relocInfo::runtime_call_type);
   ce->add_call_info_here(_info);
   ce->verify_oop_map(_info);
@@ -210,9 +210,9 @@ void MonitorExitStub::emit_code(LIR_Assembler* ce) {
   __ str(lock_reg, Address(SP));
 
   // Non-blocking leaf routine - no call info needed
-  Runtime1::StubID exit_id = ce->compilation()->has_fpu_code() ?
-                             Runtime1::monitorexit_id :
-                             Runtime1::monitorexit_nofpu_id;
+  C1StubId exit_id = ce->compilation()->has_fpu_code() ?
+                             C1StubId::monitorexit_id :
+                             C1StubId::monitorexit_nofpu_id;
   __ call(Runtime1::entry_for(exit_id), relocInfo::runtime_call_type);
   __ b(_continuation);
 }
@@ -322,10 +322,10 @@ void PatchingStub::emit_code(LIR_Assembler* ce) {
   address target = nullptr;
   relocInfo::relocType reloc_type = relocInfo::none;
   switch (_id) {
-    case access_field_id:  target = Runtime1::entry_for(Runtime1::access_field_patching_id); break;
-    case load_klass_id:    target = Runtime1::entry_for(Runtime1::load_klass_patching_id); reloc_type = relocInfo::metadata_type; break;
-    case load_mirror_id:   target = Runtime1::entry_for(Runtime1::load_mirror_patching_id); reloc_type = relocInfo::oop_type; break;
-    case load_appendix_id: target = Runtime1::entry_for(Runtime1::load_appendix_patching_id); reloc_type = relocInfo::oop_type; break;
+    case access_field_id:  target = Runtime1::entry_for(C1StubId::access_field_patching_id); break;
+    case load_klass_id:    target = Runtime1::entry_for(C1StubId::load_klass_patching_id); reloc_type = relocInfo::metadata_type; break;
+    case load_mirror_id:   target = Runtime1::entry_for(C1StubId::load_mirror_patching_id); reloc_type = relocInfo::oop_type; break;
+    case load_appendix_id: target = Runtime1::entry_for(C1StubId::load_appendix_patching_id); reloc_type = relocInfo::oop_type; break;
     default: ShouldNotReachHere();
   }
   __ bind(call_patch);
@@ -351,7 +351,7 @@ void DeoptimizeStub::emit_code(LIR_Assembler* ce) {
   __ mov_slow(Rtemp, _trap_request);
   ce->verify_reserved_argument_area_size(1);
   __ str(Rtemp, Address(SP));
-  __ call(Runtime1::entry_for(Runtime1::deoptimize_id), relocInfo::runtime_call_type);
+  __ call(Runtime1::entry_for(C1StubId::deoptimize_id), relocInfo::runtime_call_type);
   ce->add_call_info_here(_info);
   DEBUG_ONLY(__ should_not_reach_here());
 }
@@ -362,9 +362,9 @@ void ImplicitNullCheckStub::emit_code(LIR_Assembler* ce) {
   if (_info->deoptimize_on_exception()) {
     // Deoptimize, do not throw the exception, because it is
     // probably wrong to do it here.
-    a = Runtime1::entry_for(Runtime1::predicate_failed_trap_id);
+    a = Runtime1::entry_for(C1StubId::predicate_failed_trap_id);
   } else {
-    a = Runtime1::entry_for(Runtime1::throw_null_pointer_exception_id);
+    a = Runtime1::entry_for(C1StubId::throw_null_pointer_exception_id);
   }
   ce->compilation()->implicit_exception_table()->append(_offset, __ offset());
   __ bind(_entry);
diff --git a/src/hotspot/cpu/arm/c1_LIRAssembler_arm.cpp b/src/hotspot/cpu/arm/c1_LIRAssembler_arm.cpp
index 999f8fe590472..bb6a93e6f8da7 100644
--- a/src/hotspot/cpu/arm/c1_LIRAssembler_arm.cpp
+++ b/src/hotspot/cpu/arm/c1_LIRAssembler_arm.cpp
@@ -213,7 +213,7 @@ int LIR_Assembler::emit_exception_handler() {
   // check that there is really an exception
   __ verify_not_null_oop(Rexception_obj);
 
-  __ call(Runtime1::entry_for(Runtime1::handle_exception_from_callee_id), relocInfo::runtime_call_type);
+  __ call(Runtime1::entry_for(C1StubId::handle_exception_from_callee_id), relocInfo::runtime_call_type);
   __ should_not_reach_here();
 
   assert(code_offset() - offset <= exception_handler_size(), "overflow");
@@ -253,7 +253,7 @@ int LIR_Assembler::emit_unwind_handler() {
 
   // remove the activation and dispatch to the unwind handler
   __ remove_frame(initial_frame_size_in_bytes()); // restores FP and LR
-  __ jump(Runtime1::entry_for(Runtime1::unwind_exception_id), relocInfo::runtime_call_type, Rtemp);
+  __ jump(Runtime1::entry_for(C1StubId::unwind_exception_id), relocInfo::runtime_call_type, Rtemp);
 
   // Emit the slow path assembly
   if (stub != nullptr) {
@@ -1136,7 +1136,7 @@ void LIR_Assembler::emit_opTypeCheck(LIR_OpTypeCheck* op) {
       __ b(*failure_target, ne);
       // slow case
       assert(klass_RInfo == R0 && k_RInfo == R1, "runtime call setup");
-      __ call(Runtime1::entry_for(Runtime1::slow_subtype_check_id), relocInfo::runtime_call_type);
+      __ call(Runtime1::entry_for(C1StubId::slow_subtype_check_id), relocInfo::runtime_call_type);
       __ cbz(R0, *failure_target);
       if (op->should_profile()) {
         Register mdo  = klass_RInfo, recv = k_RInfo, tmp1 = Rtemp;
@@ -1210,7 +1210,7 @@ void LIR_Assembler::emit_opTypeCheck(LIR_OpTypeCheck* op) {
           __ cmp(Rtemp, k_RInfo, ne);
           __ b(*success_target, eq);
           assert(klass_RInfo == R0 && k_RInfo == R1, "runtime call setup");
-          __ call(Runtime1::entry_for(Runtime1::slow_subtype_check_id), relocInfo::runtime_call_type);
+          __ call(Runtime1::entry_for(C1StubId::slow_subtype_check_id), relocInfo::runtime_call_type);
           __ cbz(R0, *failure_target);
         }
       } else {
@@ -1227,7 +1227,7 @@ void LIR_Assembler::emit_opTypeCheck(LIR_OpTypeCheck* op) {
         __ b(*failure_target, ne);
         // slow case
         assert(klass_RInfo == R0 && k_RInfo == R1, "runtime call setup");
-        __ call(Runtime1::entry_for(Runtime1::slow_subtype_check_id), relocInfo::runtime_call_type);
+        __ call(Runtime1::entry_for(C1StubId::slow_subtype_check_id), relocInfo::runtime_call_type);
         __ cbz(R0, *failure_target);
       }
 
@@ -1303,7 +1303,7 @@ void LIR_Assembler::emit_opTypeCheck(LIR_OpTypeCheck* op) {
           }
           __ b(*success_target, eq);
           assert(klass_RInfo == R0 && k_RInfo == R1, "runtime call setup");
-          __ call(Runtime1::entry_for(Runtime1::slow_subtype_check_id), relocInfo::runtime_call_type);
+          __ call(Runtime1::entry_for(C1StubId::slow_subtype_check_id), relocInfo::runtime_call_type);
           if (!op->should_profile()) {
             move_regs(R0, res);
           } else {
@@ -1334,7 +1334,7 @@ void LIR_Assembler::emit_opTypeCheck(LIR_OpTypeCheck* op) {
         __ b(*failure_target, ne);
         // slow case
         assert(klass_RInfo == R0 && k_RInfo == R1, "runtime call setup");
-        __ call(Runtime1::entry_for(Runtime1::slow_subtype_check_id), relocInfo::runtime_call_type);
+        __ call(Runtime1::entry_for(C1StubId::slow_subtype_check_id), relocInfo::runtime_call_type);
         if (!op->should_profile()) {
           move_regs(R0, res);
         }
@@ -1981,9 +1981,9 @@ void LIR_Assembler::throw_op(LIR_Opr exceptionPC, LIR_Opr exceptionOop, CodeEmit
   assert(exceptionPC->as_register()  == Rexception_pc, "must match");
   info->add_register_oop(exceptionOop);
 
-  Runtime1::StubID handle_id = compilation()->has_fpu_code() ?
-                               Runtime1::handle_exception_id :
-                               Runtime1::handle_exception_nofpu_id;
+  C1StubId handle_id = compilation()->has_fpu_code() ?
+                               C1StubId::handle_exception_id :
+                               C1StubId::handle_exception_nofpu_id;
   Label return_address;
   __ adr(Rexception_pc, return_address);
   __ call(Runtime1::entry_for(handle_id), relocInfo::runtime_call_type);
@@ -2260,7 +2260,7 @@ void LIR_Assembler::emit_arraycopy(LIR_OpArrayCopy* op) {
       __ mov(altFP_7_11, R1);
       __ mov(R0, tmp);
       __ mov(R1, tmp2);
-      __ call(Runtime1::entry_for(Runtime1::slow_subtype_check_id), relocInfo::runtime_call_type); // does not blow any registers except R0, LR and Rtemp
+      __ call(Runtime1::entry_for(C1StubId::slow_subtype_check_id), relocInfo::runtime_call_type); // does not blow any registers except R0, LR and Rtemp
       __ cmp_32(R0, 0);
       __ mov(R0, R6);
       __ mov(R1, altFP_7_11);
diff --git a/src/hotspot/cpu/arm/c1_LIRGenerator_arm.cpp b/src/hotspot/cpu/arm/c1_LIRGenerator_arm.cpp
index f4e3812d77cff..adda0c1c290db 100644
--- a/src/hotspot/cpu/arm/c1_LIRGenerator_arm.cpp
+++ b/src/hotspot/cpu/arm/c1_LIRGenerator_arm.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2008, 2023, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2008, 2024, Oracle and/or its affiliates. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -1054,7 +1054,7 @@ void LIRGenerator::do_NewMultiArray(NewMultiArray* x) {
   args->append(rank);
   args->append(varargs);
   LIR_Opr reg = result_register_for(x->type());
-  __ call_runtime(Runtime1::entry_for(Runtime1::new_multi_array_id),
+  __ call_runtime(Runtime1::entry_for(C1StubId::new_multi_array_id),
                   LIR_OprFact::illegalOpr, reg, args, info);
 
   LIR_Opr result = rlock_result(x);
@@ -1083,7 +1083,7 @@ void LIRGenerator::do_CheckCast(CheckCast* x) {
   CodeStub* stub;
   if (x->is_incompatible_class_change_check()) {
     assert(patching_info == nullptr, "can't patch this");
-    stub = new SimpleExceptionStub(Runtime1::throw_incompatible_class_change_error_id,
+    stub = new SimpleExceptionStub(C1StubId::throw_incompatible_class_change_error_id,
                                    LIR_OprFact::illegalOpr, info_for_exception);
   } else if (x->is_invokespecial_receiver_check()) {
     assert(patching_info == nullptr, "can't patch this");
@@ -1091,7 +1091,7 @@ void LIRGenerator::do_CheckCast(CheckCast* x) {
                               Deoptimization::Reason_class_check,
                               Deoptimization::Action_none);
   } else {
-    stub = new SimpleExceptionStub(Runtime1::throw_class_cast_exception_id,
+    stub = new SimpleExceptionStub(C1StubId::throw_class_cast_exception_id,
                                    LIR_OprFact::illegalOpr, info_for_exception);
   }
 
diff --git a/src/hotspot/cpu/arm/c1_Runtime1_arm.cpp b/src/hotspot/cpu/arm/c1_Runtime1_arm.cpp
index 335baf5f16638..b5117dedc424e 100644
--- a/src/hotspot/cpu/arm/c1_Runtime1_arm.cpp
+++ b/src/hotspot/cpu/arm/c1_Runtime1_arm.cpp
@@ -65,7 +65,7 @@ int StubAssembler::call_RT(Register oop_result1, Register metadata_result, addre
   reset_last_Java_frame(Rtemp);
 
   assert(frame_size() != no_frame_size, "frame must be fixed");
-  if (_stub_id != Runtime1::forward_exception_id) {
+  if (_stub_id != (int)C1StubId::forward_exception_id) {
     ldr(R3, Address(Rthread, Thread::pending_exception_offset()));
   }
 
@@ -81,10 +81,10 @@ int StubAssembler::call_RT(Register oop_result1, Register metadata_result, addre
   // Check for pending exception
   // unpack_with_exception_in_tls path is taken through
   // Runtime1::exception_handler_for_pc
-  if (_stub_id != Runtime1::forward_exception_id) {
+  if (_stub_id != (int)C1StubId::forward_exception_id) {
     assert(frame_size() != no_frame_size, "cannot directly call forward_exception_id");
     cmp(R3, 0);
-    jump(Runtime1::entry_for(Runtime1::forward_exception_id), relocInfo::runtime_call_type, Rtemp, ne);
+    jump(Runtime1::entry_for(C1StubId::forward_exception_id), relocInfo::runtime_call_type, Rtemp, ne);
   } else {
 #ifdef ASSERT
     // Should not have pending exception in forward_exception stub
@@ -280,7 +280,7 @@ static void restore_sp_for_method_handle(StubAssembler* sasm) {
 }
 
 
-OopMapSet* Runtime1::generate_handle_exception(StubID id, StubAssembler* sasm) {
+OopMapSet* Runtime1::generate_handle_exception(C1StubId id, StubAssembler* sasm) {
   __ block_comment("generate_handle_exception");
 
   bool save_fpu_registers = false;
@@ -290,7 +290,7 @@ OopMapSet* Runtime1::generate_handle_exception(StubID id, StubAssembler* sasm) {
   OopMap* oop_map = nullptr;
 
   switch (id) {
-  case forward_exception_id: {
+  case C1StubId::forward_exception_id: {
     save_fpu_registers = HaveVFP;
     oop_map = generate_oop_map(sasm);
     __ ldr(Rexception_obj, Address(Rthread, Thread::pending_exception_offset()));
@@ -299,14 +299,14 @@ OopMapSet* Runtime1::generate_handle_exception(StubID id, StubAssembler* sasm) {
     __ str(zero, Address(Rthread, Thread::pending_exception_offset()));
     break;
   }
-  case handle_exception_id:
+  case C1StubId::handle_exception_id:
     save_fpu_registers = HaveVFP;
     // fall-through
-  case handle_exception_nofpu_id:
+  case C1StubId::handle_exception_nofpu_id:
     // At this point all registers MAY be live.
     oop_map = save_live_registers(sasm, save_fpu_registers);
     break;
-  case handle_exception_from_callee_id:
+  case C1StubId::handle_exception_from_callee_id:
     // At this point all registers except exception oop (R4/R19) and
     // exception pc (R5/R20) are dead.
     oop_map = save_live_registers(sasm);  // TODO it's not required to save all registers
@@ -328,13 +328,13 @@ OopMapSet* Runtime1::generate_handle_exception(StubID id, StubAssembler* sasm) {
   // Restore the registers that were saved at the beginning, remove
   // frame and jump to the exception handler.
   switch (id) {
-  case forward_exception_id:
-  case handle_exception_nofpu_id:
-  case handle_exception_id:
+  case C1StubId::forward_exception_id:
+  case C1StubId::handle_exception_nofpu_id:
+  case C1StubId::handle_exception_id:
     restore_live_registers(sasm, save_fpu_registers);
     // Note: the restore live registers includes the jump to LR (patched to R0)
     break;
-  case handle_exception_from_callee_id:
+  case C1StubId::handle_exception_from_callee_id:
     restore_live_registers_without_return(sasm); // must not jump immediately to handler
     restore_sp_for_method_handle(sasm);
     __ ret();
@@ -403,7 +403,7 @@ OopMapSet* Runtime1::generate_patching(StubAssembler* sasm, address target) {
 }
 
 
-OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
+OopMapSet* Runtime1::generate_code_for(C1StubId id, StubAssembler* sasm) {
   const bool must_gc_arguments = true;
   const bool dont_gc_arguments = false;
 
@@ -411,16 +411,16 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
   bool save_fpu_registers = HaveVFP;
 
   switch (id) {
-    case forward_exception_id:
+    case C1StubId::forward_exception_id:
       {
         oop_maps = generate_handle_exception(id, sasm);
         // does not return on ARM
       }
       break;
 
-    case new_instance_id:
-    case fast_new_instance_id:
-    case fast_new_instance_init_check_id:
+    case C1StubId::new_instance_id:
+    case C1StubId::fast_new_instance_id:
+    case C1StubId::fast_new_instance_init_check_id:
       {
         const Register result = R0;
         const Register klass  = R1;
@@ -436,7 +436,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case counter_overflow_id:
+    case C1StubId::counter_overflow_id:
       {
         OopMap* oop_map = save_live_registers(sasm);
         __ ldr(R1, Address(SP, arg1_offset));
@@ -448,10 +448,10 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case new_type_array_id:
-    case new_object_array_id:
+    case C1StubId::new_type_array_id:
+    case C1StubId::new_object_array_id:
       {
-        if (id == new_type_array_id) {
+        if (id == C1StubId::new_type_array_id) {
           __ set_info("new_type_array", dont_gc_arguments);
         } else {
           __ set_info("new_object_array", dont_gc_arguments);
@@ -463,7 +463,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
 
         OopMap* map = save_live_registers(sasm);
         int call_offset;
-        if (id == new_type_array_id) {
+        if (id == C1StubId::new_type_array_id) {
           call_offset = __ call_RT(result, noreg, CAST_FROM_FN_PTR(address, new_type_array), klass, length);
         } else {
           call_offset = __ call_RT(result, noreg, CAST_FROM_FN_PTR(address, new_object_array), klass, length);
@@ -477,7 +477,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case new_multi_array_id:
+    case C1StubId::new_multi_array_id:
       {
         __ set_info("new_multi_array", dont_gc_arguments);
 
@@ -500,7 +500,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case register_finalizer_id:
+    case C1StubId::register_finalizer_id:
       {
         __ set_info("register_finalizer", dont_gc_arguments);
 
@@ -521,78 +521,78 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case throw_range_check_failed_id:
+    case C1StubId::throw_range_check_failed_id:
       {
         __ set_info("range_check_failed", dont_gc_arguments);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_range_check_exception), true);
       }
       break;
 
-    case throw_index_exception_id:
+    case C1StubId::throw_index_exception_id:
       {
         __ set_info("index_range_check_failed", dont_gc_arguments);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_index_exception), true);
       }
       break;
 
-    case throw_div0_exception_id:
+    case C1StubId::throw_div0_exception_id:
       {
         __ set_info("throw_div0_exception", dont_gc_arguments);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_div0_exception), false);
       }
       break;
 
-    case throw_null_pointer_exception_id:
+    case C1StubId::throw_null_pointer_exception_id:
       {
         __ set_info("throw_null_pointer_exception", dont_gc_arguments);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_null_pointer_exception), false);
       }
       break;
 
-    case handle_exception_nofpu_id:
-    case handle_exception_id:
+    case C1StubId::handle_exception_nofpu_id:
+    case C1StubId::handle_exception_id:
       {
         __ set_info("handle_exception", dont_gc_arguments);
         oop_maps = generate_handle_exception(id, sasm);
       }
       break;
 
-    case handle_exception_from_callee_id:
+    case C1StubId::handle_exception_from_callee_id:
       {
         __ set_info("handle_exception_from_callee", dont_gc_arguments);
         oop_maps = generate_handle_exception(id, sasm);
       }
       break;
 
-    case unwind_exception_id:
+    case C1StubId::unwind_exception_id:
       {
         __ set_info("unwind_exception", dont_gc_arguments);
         generate_unwind_exception(sasm);
       }
       break;
 
-    case throw_array_store_exception_id:
+    case C1StubId::throw_array_store_exception_id:
       {
         __ set_info("throw_array_store_exception", dont_gc_arguments);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_array_store_exception), true);
       }
       break;
 
-    case throw_class_cast_exception_id:
+    case C1StubId::throw_class_cast_exception_id:
       {
         __ set_info("throw_class_cast_exception", dont_gc_arguments);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_class_cast_exception), true);
       }
       break;
 
-    case throw_incompatible_class_change_error_id:
+    case C1StubId::throw_incompatible_class_change_error_id:
       {
         __ set_info("throw_incompatible_class_cast_exception", dont_gc_arguments);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_incompatible_class_change_error), false);
       }
       break;
 
-    case slow_subtype_check_id:
+    case C1StubId::slow_subtype_check_id:
       {
         // (in)  R0 - sub, destroyed,
         // (in)  R1 - super, not changed
@@ -625,10 +625,10 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case monitorenter_nofpu_id:
+    case C1StubId::monitorenter_nofpu_id:
       save_fpu_registers = false;
       // fall through
-    case monitorenter_id:
+    case C1StubId::monitorenter_id:
       {
         __ set_info("monitorenter", dont_gc_arguments);
         const Register obj  = R1;
@@ -643,10 +643,10 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case monitorexit_nofpu_id:
+    case C1StubId::monitorexit_nofpu_id:
       save_fpu_registers = false;
       // fall through
-    case monitorexit_id:
+    case C1StubId::monitorexit_id:
       {
         __ set_info("monitorexit", dont_gc_arguments);
         const Register lock = R1;
@@ -659,7 +659,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case deoptimize_id:
+    case C1StubId::deoptimize_id:
       {
         __ set_info("deoptimize", dont_gc_arguments);
         OopMap* oop_map = save_live_registers(sasm);
@@ -675,35 +675,35 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case access_field_patching_id:
+    case C1StubId::access_field_patching_id:
       {
         __ set_info("access_field_patching", dont_gc_arguments);
         oop_maps = generate_patching(sasm, CAST_FROM_FN_PTR(address, access_field_patching));
       }
       break;
 
-    case load_klass_patching_id:
+    case C1StubId::load_klass_patching_id:
       {
         __ set_info("load_klass_patching", dont_gc_arguments);
         oop_maps = generate_patching(sasm, CAST_FROM_FN_PTR(address, move_klass_patching));
       }
       break;
 
-    case load_appendix_patching_id:
+    case C1StubId::load_appendix_patching_id:
       {
         __ set_info("load_appendix_patching", dont_gc_arguments);
         oop_maps = generate_patching(sasm, CAST_FROM_FN_PTR(address, move_appendix_patching));
       }
       break;
 
-    case load_mirror_patching_id:
+    case C1StubId::load_mirror_patching_id:
       {
         __ set_info("load_mirror_patching", dont_gc_arguments);
         oop_maps = generate_patching(sasm, CAST_FROM_FN_PTR(address, move_mirror_patching));
       }
       break;
 
-    case predicate_failed_trap_id:
+    case C1StubId::predicate_failed_trap_id:
       {
         __ set_info("predicate_failed_trap", dont_gc_arguments);
 
diff --git a/src/hotspot/cpu/ppc/c1_CodeStubs_ppc.cpp b/src/hotspot/cpu/ppc/c1_CodeStubs_ppc.cpp
index dc70c73d4b330..451f3b7e9cd6b 100644
--- a/src/hotspot/cpu/ppc/c1_CodeStubs_ppc.cpp
+++ b/src/hotspot/cpu/ppc/c1_CodeStubs_ppc.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999, 2023, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 1999, 2024, Oracle and/or its affiliates. All rights reserved.
  * Copyright (c) 2012, 2021 SAP SE. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
@@ -68,7 +68,7 @@ void RangeCheckStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
 
   if (_info->deoptimize_on_exception()) {
-    address a = Runtime1::entry_for(Runtime1::predicate_failed_trap_id);
+    address a = Runtime1::entry_for(C1StubId::predicate_failed_trap_id);
     //__ load_const_optimized(R0, a);
     __ add_const_optimized(R0, R29_TOC, MacroAssembler::offset_to_global_toc(a));
     __ mtctr(R0);
@@ -79,8 +79,8 @@ void RangeCheckStub::emit_code(LIR_Assembler* ce) {
     return;
   }
 
-  address stub = _throw_index_out_of_bounds_exception ? Runtime1::entry_for(Runtime1::throw_index_exception_id)
-                                                      : Runtime1::entry_for(Runtime1::throw_range_check_failed_id);
+  address stub = _throw_index_out_of_bounds_exception ? Runtime1::entry_for(C1StubId::throw_index_exception_id)
+                                                      : Runtime1::entry_for(C1StubId::throw_range_check_failed_id);
   //__ load_const_optimized(R0, stub);
   __ add_const_optimized(R0, R29_TOC, MacroAssembler::offset_to_global_toc(stub));
   __ mtctr(R0);
@@ -109,7 +109,7 @@ PredicateFailedStub::PredicateFailedStub(CodeEmitInfo* info) {
 
 void PredicateFailedStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
-  address a = Runtime1::entry_for(Runtime1::predicate_failed_trap_id);
+  address a = Runtime1::entry_for(C1StubId::predicate_failed_trap_id);
   //__ load_const_optimized(R0, a);
   __ add_const_optimized(R0, R29_TOC, MacroAssembler::offset_to_global_toc(a));
   __ mtctr(R0);
@@ -133,7 +133,7 @@ void CounterOverflowStub::emit_code(LIR_Assembler* ce) {
   __ load_const_optimized(R0, md.value());
   __ std(R0, -8, R1_SP);
 
-  address a = Runtime1::entry_for(Runtime1::counter_overflow_id);
+  address a = Runtime1::entry_for(C1StubId::counter_overflow_id);
   //__ load_const_optimized(R0, a);
   __ add_const_optimized(R0, R29_TOC, MacroAssembler::offset_to_global_toc(a));
   __ mtctr(R0);
@@ -150,7 +150,7 @@ void DivByZeroStub::emit_code(LIR_Assembler* ce) {
     ce->compilation()->implicit_exception_table()->append(_offset, __ offset());
   }
   __ bind(_entry);
-  address stub = Runtime1::entry_for(Runtime1::throw_div0_exception_id);
+  address stub = Runtime1::entry_for(C1StubId::throw_div0_exception_id);
   //__ load_const_optimized(R0, stub);
   __ add_const_optimized(R0, R29_TOC, MacroAssembler::offset_to_global_toc(stub));
   __ mtctr(R0);
@@ -165,9 +165,9 @@ void ImplicitNullCheckStub::emit_code(LIR_Assembler* ce) {
   address a;
   if (_info->deoptimize_on_exception()) {
     // Deoptimize, do not throw the exception, because it is probably wrong to do it here.
-    a = Runtime1::entry_for(Runtime1::predicate_failed_trap_id);
+    a = Runtime1::entry_for(C1StubId::predicate_failed_trap_id);
   } else {
-    a = Runtime1::entry_for(Runtime1::throw_null_pointer_exception_id);
+    a = Runtime1::entry_for(C1StubId::throw_null_pointer_exception_id);
   }
 
   if (ImplicitNullChecks || TrapBasedNullChecks) {
@@ -199,14 +199,14 @@ void SimpleExceptionStub::emit_code(LIR_Assembler* ce) {
 
 
 // Implementation of NewInstanceStub
-NewInstanceStub::NewInstanceStub(LIR_Opr klass_reg, LIR_Opr result, ciInstanceKlass* klass, CodeEmitInfo* info, Runtime1::StubID stub_id) {
+NewInstanceStub::NewInstanceStub(LIR_Opr klass_reg, LIR_Opr result, ciInstanceKlass* klass, CodeEmitInfo* info, C1StubId stub_id) {
   _result = result;
   _klass = klass;
   _klass_reg = klass_reg;
   _info = new CodeEmitInfo(info);
-  assert(stub_id == Runtime1::new_instance_id                 ||
-         stub_id == Runtime1::fast_new_instance_id            ||
-         stub_id == Runtime1::fast_new_instance_init_check_id,
+  assert(stub_id == C1StubId::new_instance_id                 ||
+         stub_id == C1StubId::fast_new_instance_id            ||
+         stub_id == C1StubId::fast_new_instance_init_check_id,
          "need new_instance id");
   _stub_id = stub_id;
 }
@@ -236,7 +236,7 @@ NewTypeArrayStub::NewTypeArrayStub(LIR_Opr klass_reg, LIR_Opr length, LIR_Opr re
 void NewTypeArrayStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
 
-  address entry = Runtime1::entry_for(Runtime1::new_type_array_id);
+  address entry = Runtime1::entry_for(C1StubId::new_type_array_id);
   //__ load_const_optimized(R0, entry);
   __ add_const_optimized(R0, R29_TOC, MacroAssembler::offset_to_global_toc(entry));
   __ mr_if_needed(/*op->tmp1()->as_register()*/ R5_ARG3, _length->as_register()); // already sign-extended
@@ -259,7 +259,7 @@ NewObjectArrayStub::NewObjectArrayStub(LIR_Opr klass_reg, LIR_Opr length, LIR_Op
 void NewObjectArrayStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
 
-  address entry = Runtime1::entry_for(Runtime1::new_object_array_id);
+  address entry = Runtime1::entry_for(C1StubId::new_object_array_id);
   //__ load_const_optimized(R0, entry);
   __ add_const_optimized(R0, R29_TOC, MacroAssembler::offset_to_global_toc(entry));
   __ mr_if_needed(/*op->tmp1()->as_register()*/ R5_ARG3, _length->as_register()); // already sign-extended
@@ -272,7 +272,7 @@ void NewObjectArrayStub::emit_code(LIR_Assembler* ce) {
 
 void MonitorEnterStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
-  address stub = Runtime1::entry_for(ce->compilation()->has_fpu_code() ? Runtime1::monitorenter_id : Runtime1::monitorenter_nofpu_id);
+  address stub = Runtime1::entry_for(ce->compilation()->has_fpu_code() ? C1StubId::monitorenter_id : C1StubId::monitorenter_nofpu_id);
   //__ load_const_optimized(R0, stub);
   __ add_const_optimized(R0, R29_TOC, MacroAssembler::offset_to_global_toc(stub));
   __ mr_if_needed(/*scratch_opr()->as_register()*/ R4_ARG2, _obj_reg->as_register());
@@ -289,7 +289,7 @@ void MonitorExitStub::emit_code(LIR_Assembler* ce) {
   if (_compute_lock) {
     ce->monitor_address(_monitor_ix, _lock_reg);
   }
-  address stub = Runtime1::entry_for(ce->compilation()->has_fpu_code() ? Runtime1::monitorexit_id : Runtime1::monitorexit_nofpu_id);
+  address stub = Runtime1::entry_for(ce->compilation()->has_fpu_code() ? C1StubId::monitorexit_id : C1StubId::monitorexit_nofpu_id);
   //__ load_const_optimized(R0, stub);
   __ add_const_optimized(R0, R29_TOC, MacroAssembler::offset_to_global_toc(stub));
   assert(_lock_reg->as_register() == R4_ARG2, "");
@@ -403,12 +403,12 @@ void PatchingStub::emit_code(LIR_Assembler* ce) {
   address target = nullptr;
   relocInfo::relocType reloc_type = relocInfo::none;
   switch (_id) {
-    case access_field_id:  target = Runtime1::entry_for(Runtime1::access_field_patching_id); break;
-    case load_klass_id:    target = Runtime1::entry_for(Runtime1::load_klass_patching_id);
+    case access_field_id:  target = Runtime1::entry_for(C1StubId::access_field_patching_id); break;
+    case load_klass_id:    target = Runtime1::entry_for(C1StubId::load_klass_patching_id);
                            reloc_type = relocInfo::metadata_type; break;
-    case load_mirror_id:   target = Runtime1::entry_for(Runtime1::load_mirror_patching_id);
+    case load_mirror_id:   target = Runtime1::entry_for(C1StubId::load_mirror_patching_id);
                            reloc_type = relocInfo::oop_type; break;
-    case load_appendix_id: target = Runtime1::entry_for(Runtime1::load_appendix_patching_id);
+    case load_appendix_id: target = Runtime1::entry_for(C1StubId::load_appendix_patching_id);
                            reloc_type = relocInfo::oop_type; break;
     default: ShouldNotReachHere();
   }
@@ -434,7 +434,7 @@ void PatchingStub::emit_code(LIR_Assembler* ce) {
 
 void DeoptimizeStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
-  address stub = Runtime1::entry_for(Runtime1::deoptimize_id);
+  address stub = Runtime1::entry_for(C1StubId::deoptimize_id);
   //__ load_const_optimized(R0, stub);
   __ add_const_optimized(R0, R29_TOC, MacroAssembler::offset_to_global_toc(stub));
   __ mtctr(R0);
diff --git a/src/hotspot/cpu/ppc/c1_LIRAssembler_ppc.cpp b/src/hotspot/cpu/ppc/c1_LIRAssembler_ppc.cpp
index 2191b894f6e16..42934dc7c3179 100644
--- a/src/hotspot/cpu/ppc/c1_LIRAssembler_ppc.cpp
+++ b/src/hotspot/cpu/ppc/c1_LIRAssembler_ppc.cpp
@@ -176,7 +176,7 @@ int LIR_Assembler::emit_exception_handler() {
   }
 
   int offset = code_offset();
-  address entry_point = CAST_FROM_FN_PTR(address, Runtime1::entry_for(Runtime1::handle_exception_from_callee_id));
+  address entry_point = CAST_FROM_FN_PTR(address, Runtime1::entry_for(C1StubId::handle_exception_from_callee_id));
   //__ load_const_optimized(R0, entry_point);
   __ add_const_optimized(R0, R29_TOC, MacroAssembler::offset_to_global_toc(entry_point));
   __ mtctr(R0);
@@ -222,7 +222,7 @@ int LIR_Assembler::emit_unwind_handler() {
   }
 
   // Dispatch to the unwind logic.
-  address unwind_stub = Runtime1::entry_for(Runtime1::unwind_exception_id);
+  address unwind_stub = Runtime1::entry_for(C1StubId::unwind_exception_id);
   //__ load_const_optimized(R0, unwind_stub);
   __ add_const_optimized(R0, R29_TOC, MacroAssembler::offset_to_global_toc(unwind_stub));
   if (preserve_exception) { __ mr(Rexception, Rexception_save); }
@@ -1800,8 +1800,8 @@ void LIR_Assembler::throw_op(LIR_Opr exceptionPC, LIR_Opr exceptionOop, CodeEmit
   __ calculate_address_from_global_toc(exceptionPC->as_register(), pc_for_athrow, true, true, /*add_relocation*/ true);
   add_call_info(pc_for_athrow_offset, info); // for exception handler
 
-  address stub = Runtime1::entry_for(compilation()->has_fpu_code() ? Runtime1::handle_exception_id
-                                                                   : Runtime1::handle_exception_nofpu_id);
+  address stub = Runtime1::entry_for(compilation()->has_fpu_code() ? C1StubId::handle_exception_id
+                                                                   : C1StubId::handle_exception_nofpu_id);
   //__ load_const_optimized(R0, stub);
   __ add_const_optimized(R0, R29_TOC, MacroAssembler::offset_to_global_toc(stub));
   __ mtctr(R0);
@@ -2001,7 +2001,7 @@ void LIR_Assembler::emit_arraycopy(LIR_OpArrayCopy* op) {
       __ check_klass_subtype_fast_path(sub_klass, super_klass, tmp, tmp2,
                                        &cont, copyfunc_addr != nullptr ? &copyfunc : &slow, nullptr);
 
-      address slow_stc = Runtime1::entry_for(Runtime1::slow_subtype_check_id);
+      address slow_stc = Runtime1::entry_for(C1StubId::slow_subtype_check_id);
       //__ load_const_optimized(tmp, slow_stc, tmp2);
       __ calculate_address_from_global_toc(tmp, slow_stc, true, true, false);
       __ mtctr(tmp);
@@ -2452,7 +2452,7 @@ void LIR_Assembler::emit_typecheck_helper(LIR_OpTypeCheck *op, Label* success, L
       __ b(*success);
     } else {
       // Call out-of-line instance of __ check_klass_subtype_slow_path(...):
-      address entry = Runtime1::entry_for(Runtime1::slow_subtype_check_id);
+      address entry = Runtime1::entry_for(C1StubId::slow_subtype_check_id);
       // Stub needs fixed registers (tmp1-3).
       Register original_k_RInfo = op->tmp1()->as_register();
       Register original_klass_RInfo = op->tmp2()->as_register();
@@ -2543,7 +2543,7 @@ void LIR_Assembler::emit_opTypeCheck(LIR_OpTypeCheck* op) {
     __ check_klass_subtype_fast_path(klass_RInfo, k_RInfo, Rtmp1, R0, &done, &failure, nullptr);
 
     // Call out-of-line instance of __ check_klass_subtype_slow_path(...):
-    const address slow_path = Runtime1::entry_for(Runtime1::slow_subtype_check_id);
+    const address slow_path = Runtime1::entry_for(C1StubId::slow_subtype_check_id);
     //__ load_const_optimized(R0, slow_path);
     __ add_const_optimized(R0, R29_TOC, MacroAssembler::offset_to_global_toc(slow_path));
     __ mtctr(R0);
@@ -2850,8 +2850,8 @@ void LIR_Assembler::negate(LIR_Opr left, LIR_Opr dest, LIR_Opr tmp) {
 void LIR_Assembler::rt_call(LIR_Opr result, address dest,
                             const LIR_OprList* args, LIR_Opr tmp, CodeEmitInfo* info) {
   // Stubs: Called via rt_call, but dest is a stub address (no function descriptor).
-  if (dest == Runtime1::entry_for(Runtime1::register_finalizer_id) ||
-      dest == Runtime1::entry_for(Runtime1::new_multi_array_id   )) {
+  if (dest == Runtime1::entry_for(C1StubId::register_finalizer_id) ||
+      dest == Runtime1::entry_for(C1StubId::new_multi_array_id   )) {
     //__ load_const_optimized(R0, dest);
     __ add_const_optimized(R0, R29_TOC, MacroAssembler::offset_to_global_toc(dest));
     __ mtctr(R0);
diff --git a/src/hotspot/cpu/ppc/c1_LIRGenerator_ppc.cpp b/src/hotspot/cpu/ppc/c1_LIRGenerator_ppc.cpp
index 04762a22c6110..7973e9d05459e 100644
--- a/src/hotspot/cpu/ppc/c1_LIRGenerator_ppc.cpp
+++ b/src/hotspot/cpu/ppc/c1_LIRGenerator_ppc.cpp
@@ -1032,7 +1032,7 @@ void LIRGenerator::do_NewMultiArray(NewMultiArray* x) {
   args->append(rank);
   args->append(varargs);
   const LIR_Opr reg = result_register_for(x->type());
-  __ call_runtime(Runtime1::entry_for(Runtime1::new_multi_array_id),
+  __ call_runtime(Runtime1::entry_for(C1StubId::new_multi_array_id),
                   LIR_OprFact::illegalOpr,
                   reg, args, info);
 
@@ -1067,7 +1067,7 @@ void LIRGenerator::do_CheckCast(CheckCast* x) {
 
   if (x->is_incompatible_class_change_check()) {
     assert(patching_info == nullptr, "can't patch this");
-    stub = new SimpleExceptionStub(Runtime1::throw_incompatible_class_change_error_id,
+    stub = new SimpleExceptionStub(C1StubId::throw_incompatible_class_change_error_id,
                                    LIR_OprFact::illegalOpr, info_for_exception);
   } else if (x->is_invokespecial_receiver_check()) {
     assert(patching_info == nullptr, "can't patch this");
@@ -1075,7 +1075,7 @@ void LIRGenerator::do_CheckCast(CheckCast* x) {
                               Deoptimization::Reason_class_check,
                               Deoptimization::Action_none);
   } else {
-    stub = new SimpleExceptionStub(Runtime1::throw_class_cast_exception_id, obj.result(), info_for_exception);
+    stub = new SimpleExceptionStub(C1StubId::throw_class_cast_exception_id, obj.result(), info_for_exception);
   }
   // Following registers are used by slow_subtype_check:
   LIR_Opr tmp1 = FrameMap::R4_oop_opr; // super_klass
diff --git a/src/hotspot/cpu/ppc/c1_MacroAssembler_ppc.cpp b/src/hotspot/cpu/ppc/c1_MacroAssembler_ppc.cpp
index 059bb2eae0c3a..c05e97a4e9aa3 100644
--- a/src/hotspot/cpu/ppc/c1_MacroAssembler_ppc.cpp
+++ b/src/hotspot/cpu/ppc/c1_MacroAssembler_ppc.cpp
@@ -293,7 +293,7 @@ void C1_MacroAssembler::initialize_object(
   if (CURRENT_ENV->dtrace_alloc_probes()) {
     Unimplemented();
 //    assert(obj == O0, "must be");
-//    call(CAST_FROM_FN_PTR(address, Runtime1::entry_for(Runtime1::dtrace_object_alloc_id)),
+//    call(CAST_FROM_FN_PTR(address, Runtime1::entry_for(C1StubId::dtrace_object_alloc_id)),
 //         relocInfo::runtime_call_type);
   }
 
@@ -369,7 +369,7 @@ void C1_MacroAssembler::allocate_array(
   if (CURRENT_ENV->dtrace_alloc_probes()) {
     Unimplemented();
     //assert(obj == O0, "must be");
-    //call(CAST_FROM_FN_PTR(address, Runtime1::entry_for(Runtime1::dtrace_object_alloc_id)),
+    //call(CAST_FROM_FN_PTR(address, Runtime1::entry_for(C1StubId::dtrace_object_alloc_id)),
     //     relocInfo::runtime_call_type);
   }
 
@@ -398,7 +398,7 @@ void C1_MacroAssembler::null_check(Register r, Label* Lnull) {
   if (TrapBasedNullChecks) { // SIGTRAP based
     trap_null_check(r);
   } else { // explicit
-    //const address exception_entry = Runtime1::entry_for(Runtime1::throw_null_pointer_exception_id);
+    //const address exception_entry = Runtime1::entry_for(C1StubId::throw_null_pointer_exception_id);
     assert(Lnull != nullptr, "must have Label for explicit check");
     cmpdi(CCR0, r, 0);
     bc_far_optimized(Assembler::bcondCRbiIs1, bi0(CCR0, Assembler::equal), *Lnull);
diff --git a/src/hotspot/cpu/ppc/c1_Runtime1_ppc.cpp b/src/hotspot/cpu/ppc/c1_Runtime1_ppc.cpp
index adddfda4ee74f..654626d66d812 100644
--- a/src/hotspot/cpu/ppc/c1_Runtime1_ppc.cpp
+++ b/src/hotspot/cpu/ppc/c1_Runtime1_ppc.cpp
@@ -97,12 +97,12 @@ int StubAssembler::call_RT(Register oop_result1, Register metadata_result,
       //load_const_optimized(R0, StubRoutines::forward_exception_entry());
       //mtctr(R0);
       //bctr();
-    } else if (_stub_id == Runtime1::forward_exception_id) {
+    } else if (_stub_id == (int)C1StubId::forward_exception_id) {
       should_not_reach_here();
     } else {
       // keep stub frame for next call_RT
-      //load_const_optimized(R0, Runtime1::entry_for(Runtime1::forward_exception_id));
-      add_const_optimized(R0, R29_TOC, MacroAssembler::offset_to_global_toc(Runtime1::entry_for(Runtime1::forward_exception_id)));
+      //load_const_optimized(R0, Runtime1::entry_for(C1StubId::forward_exception_id));
+      add_const_optimized(R0, R29_TOC, MacroAssembler::offset_to_global_toc(Runtime1::entry_for(C1StubId::forward_exception_id)));
       mtctr(R0);
       bctr();
     }
@@ -388,7 +388,7 @@ OopMapSet* Runtime1::generate_patching(StubAssembler* sasm, address target) {
   return oop_maps;
 }
 
-OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
+OopMapSet* Runtime1::generate_code_for(C1StubId id, StubAssembler* sasm) {
   OopMapSet* oop_maps = nullptr;
 
   // For better readability.
@@ -397,22 +397,22 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
 
   // Stub code & info for the different stubs.
   switch (id) {
-    case forward_exception_id:
+    case C1StubId::forward_exception_id:
       {
         oop_maps = generate_handle_exception(id, sasm);
       }
       break;
 
-    case new_instance_id:
-    case fast_new_instance_id:
-    case fast_new_instance_init_check_id:
+    case C1StubId::new_instance_id:
+    case C1StubId::fast_new_instance_id:
+    case C1StubId::fast_new_instance_init_check_id:
       {
-        if (id == new_instance_id) {
+        if (id == C1StubId::new_instance_id) {
           __ set_info("new_instance", dont_gc_arguments);
-        } else if (id == fast_new_instance_id) {
+        } else if (id == C1StubId::fast_new_instance_id) {
           __ set_info("fast new_instance", dont_gc_arguments);
         } else {
-          assert(id == fast_new_instance_init_check_id, "bad StubID");
+          assert(id == C1StubId::fast_new_instance_init_check_id, "bad C1StubId");
           __ set_info("fast new_instance init check", dont_gc_arguments);
         }
 
@@ -422,15 +422,15 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case counter_overflow_id:
+    case C1StubId::counter_overflow_id:
         // Bci and method are on stack.
         oop_maps = stub_call_with_stack_parms(sasm, noreg, CAST_FROM_FN_PTR(address, counter_overflow), 2);
       break;
 
-    case new_type_array_id:
-    case new_object_array_id:
+    case C1StubId::new_type_array_id:
+    case C1StubId::new_object_array_id:
       {
-        if (id == new_type_array_id) {
+        if (id == C1StubId::new_type_array_id) {
           __ set_info("new_type_array", dont_gc_arguments);
         } else {
           __ set_info("new_object_array", dont_gc_arguments);
@@ -439,7 +439,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
 #ifdef ASSERT
         // Assert object type is really an array of the proper kind.
         {
-          int tag = (id == new_type_array_id) ? Klass::_lh_array_tag_type_value : Klass::_lh_array_tag_obj_value;
+          int tag = (id == C1StubId::new_type_array_id) ? Klass::_lh_array_tag_type_value : Klass::_lh_array_tag_obj_value;
           Label ok;
           __ lwz(R0, in_bytes(Klass::layout_helper_offset()), R4_ARG2);
           __ srawi(R0, R0, Klass::_lh_array_tag_shift);
@@ -453,7 +453,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
 
         // We don't support eden allocation.
 
-        if (id == new_type_array_id) {
+        if (id == C1StubId::new_type_array_id) {
           oop_maps = generate_stub_call(sasm, R3_RET, CAST_FROM_FN_PTR(address, new_type_array), R4_ARG2, R5_ARG3);
         } else {
           oop_maps = generate_stub_call(sasm, R3_RET, CAST_FROM_FN_PTR(address, new_object_array), R4_ARG2, R5_ARG3);
@@ -461,7 +461,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case new_multi_array_id:
+    case C1StubId::new_multi_array_id:
       {
         // R4: klass
         // R5: rank
@@ -471,7 +471,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case register_finalizer_id:
+    case C1StubId::register_finalizer_id:
       {
         __ set_info("register_finalizer", dont_gc_arguments);
         // This code is called via rt_call. Hence, caller-save registers have been saved.
@@ -501,50 +501,50 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case throw_range_check_failed_id:
+    case C1StubId::throw_range_check_failed_id:
       {
         __ set_info("range_check_failed", dont_gc_arguments); // Arguments will be discarded.
         oop_maps = generate_exception_throw_with_stack_parms(sasm, CAST_FROM_FN_PTR(address, throw_range_check_exception), 2);
       }
       break;
 
-    case throw_index_exception_id:
+    case C1StubId::throw_index_exception_id:
       {
         __ set_info("index_range_check_failed", dont_gc_arguments); // Arguments will be discarded.
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_index_exception), true);
       }
       break;
 
-    case throw_div0_exception_id:
+    case C1StubId::throw_div0_exception_id:
       {
         __ set_info("throw_div0_exception", dont_gc_arguments);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_div0_exception), false);
       }
       break;
 
-    case throw_null_pointer_exception_id:
+    case C1StubId::throw_null_pointer_exception_id:
       {
         __ set_info("throw_null_pointer_exception", dont_gc_arguments);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_null_pointer_exception), false);
       }
       break;
 
-    case handle_exception_nofpu_id:
-    case handle_exception_id:
+    case C1StubId::handle_exception_nofpu_id:
+    case C1StubId::handle_exception_id:
       {
         __ set_info("handle_exception", dont_gc_arguments);
         oop_maps = generate_handle_exception(id, sasm);
       }
       break;
 
-    case handle_exception_from_callee_id:
+    case C1StubId::handle_exception_from_callee_id:
       {
         __ set_info("handle_exception_from_callee", dont_gc_arguments);
         oop_maps = generate_handle_exception(id, sasm);
       }
       break;
 
-    case unwind_exception_id:
+    case C1StubId::unwind_exception_id:
       {
         const Register Rexception    = R3 /*LIRGenerator::exceptionOopOpr()*/,
                        Rexception_pc = R4 /*LIRGenerator::exceptionPcOpr()*/,
@@ -572,28 +572,28 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case throw_array_store_exception_id:
+    case C1StubId::throw_array_store_exception_id:
       {
         __ set_info("throw_array_store_exception", dont_gc_arguments);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_array_store_exception), true);
       }
       break;
 
-    case throw_class_cast_exception_id:
+    case C1StubId::throw_class_cast_exception_id:
       {
         __ set_info("throw_class_cast_exception", dont_gc_arguments);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_class_cast_exception), true);
       }
       break;
 
-    case throw_incompatible_class_change_error_id:
+    case C1StubId::throw_incompatible_class_change_error_id:
       {
         __ set_info("throw_incompatible_class_cast_exception", dont_gc_arguments);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_incompatible_class_change_error), false);
       }
       break;
 
-    case slow_subtype_check_id:
+    case C1StubId::slow_subtype_check_id:
       { // Support for uint StubRoutine::partial_subtype_check( Klass sub, Klass super );
         const Register sub_klass = R5,
                        super_klass = R4,
@@ -605,12 +605,12 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case monitorenter_nofpu_id:
-    case monitorenter_id:
+    case C1StubId::monitorenter_nofpu_id:
+    case C1StubId::monitorenter_id:
       {
         __ set_info("monitorenter", dont_gc_arguments);
 
-        int save_fpu_registers = (id == monitorenter_id);
+        int save_fpu_registers = (id == C1StubId::monitorenter_id);
         // Make a frame and preserve the caller's caller-save registers.
         OopMap* oop_map = save_live_registers(sasm, save_fpu_registers);
 
@@ -624,15 +624,15 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case monitorexit_nofpu_id:
-    case monitorexit_id:
+    case C1StubId::monitorexit_nofpu_id:
+    case C1StubId::monitorexit_id:
       {
         // note: Really a leaf routine but must setup last java sp
         //       => use call_RT for now (speed can be improved by
         //       doing last java sp setup manually).
         __ set_info("monitorexit", dont_gc_arguments);
 
-        int save_fpu_registers = (id == monitorexit_id);
+        int save_fpu_registers = (id == C1StubId::monitorexit_id);
         // Make a frame and preserve the caller's caller-save registers.
         OopMap* oop_map = save_live_registers(sasm, save_fpu_registers);
 
@@ -646,7 +646,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case deoptimize_id:
+    case C1StubId::deoptimize_id:
       {
         __ set_info("deoptimize", dont_gc_arguments);
         __ std(R0, -8, R1_SP); // Pass trap_request on stack.
@@ -662,35 +662,35 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case access_field_patching_id:
+    case C1StubId::access_field_patching_id:
       {
         __ set_info("access_field_patching", dont_gc_arguments);
         oop_maps = generate_patching(sasm, CAST_FROM_FN_PTR(address, access_field_patching));
       }
       break;
 
-    case load_klass_patching_id:
+    case C1StubId::load_klass_patching_id:
       {
         __ set_info("load_klass_patching", dont_gc_arguments);
         oop_maps = generate_patching(sasm, CAST_FROM_FN_PTR(address, move_klass_patching));
       }
       break;
 
-    case load_mirror_patching_id:
+    case C1StubId::load_mirror_patching_id:
       {
         __ set_info("load_mirror_patching", dont_gc_arguments);
         oop_maps = generate_patching(sasm, CAST_FROM_FN_PTR(address, move_mirror_patching));
       }
       break;
 
-    case load_appendix_patching_id:
+    case C1StubId::load_appendix_patching_id:
       {
         __ set_info("load_appendix_patching", dont_gc_arguments);
         oop_maps = generate_patching(sasm, CAST_FROM_FN_PTR(address, move_appendix_patching));
       }
       break;
 
-    case dtrace_object_alloc_id:
+    case C1StubId::dtrace_object_alloc_id:
       { // O0: object
         __ unimplemented("stub dtrace_object_alloc_id");
         __ set_info("dtrace_object_alloc", dont_gc_arguments);
@@ -710,7 +710,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case predicate_failed_trap_id:
+    case C1StubId::predicate_failed_trap_id:
       {
         __ set_info("predicate_failed_trap", dont_gc_arguments);
         OopMap* oop_map = save_live_registers(sasm);
@@ -754,7 +754,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
 }
 
 
-OopMapSet* Runtime1::generate_handle_exception(StubID id, StubAssembler* sasm) {
+OopMapSet* Runtime1::generate_handle_exception(C1StubId id, StubAssembler* sasm) {
   __ block_comment("generate_handle_exception");
 
   // Save registers, if required.
@@ -764,7 +764,7 @@ OopMapSet* Runtime1::generate_handle_exception(StubID id, StubAssembler* sasm) {
                  Rexception_pc = R4 /*LIRGenerator::exceptionPcOpr()*/;
 
   switch (id) {
-  case forward_exception_id:
+  case C1StubId::forward_exception_id:
     // We're handling an exception in the context of a compiled frame.
     // The registers have been saved in the standard places. Perform
     // an exception lookup in the caller and dispatch to the handler
@@ -780,12 +780,12 @@ OopMapSet* Runtime1::generate_handle_exception(StubID id, StubAssembler* sasm) {
     __ ld(Rexception_pc, _abi0(lr), Rexception_pc);
     __ std(R0, in_bytes(JavaThread::pending_exception_offset()), R16_thread);
     break;
-  case handle_exception_nofpu_id:
-  case handle_exception_id:
+  case C1StubId::handle_exception_nofpu_id:
+  case C1StubId::handle_exception_id:
     // At this point all registers MAY be live.
-    oop_map = save_live_registers(sasm, id != handle_exception_nofpu_id, Rexception_pc);
+    oop_map = save_live_registers(sasm, id != C1StubId::handle_exception_nofpu_id, Rexception_pc);
     break;
-  case handle_exception_from_callee_id:
+  case C1StubId::handle_exception_from_callee_id:
     // At this point all registers except exception oop and exception pc are dead.
     oop_map = new OopMap(frame_size_in_bytes / sizeof(jint), 0);
     sasm->set_frame_size(frame_size_in_bytes / BytesPerWord);
@@ -824,13 +824,13 @@ OopMapSet* Runtime1::generate_handle_exception(StubID id, StubAssembler* sasm) {
   // Restore the registers that were saved at the beginning, remove
   // the frame and jump to the exception handler.
   switch (id) {
-  case forward_exception_id:
-  case handle_exception_nofpu_id:
-  case handle_exception_id:
-    restore_live_registers(sasm, noreg, noreg, id != handle_exception_nofpu_id);
+  case C1StubId::forward_exception_id:
+  case C1StubId::handle_exception_nofpu_id:
+  case C1StubId::handle_exception_id:
+    restore_live_registers(sasm, noreg, noreg, id != C1StubId::handle_exception_nofpu_id);
     __ bctr();
     break;
-  case handle_exception_from_callee_id: {
+  case C1StubId::handle_exception_from_callee_id: {
     __ pop_frame();
     __ ld(Rexception_pc, _abi0(lr), R1_SP);
     __ mtlr(Rexception_pc);
diff --git a/src/hotspot/cpu/riscv/assembler_riscv.hpp b/src/hotspot/cpu/riscv/assembler_riscv.hpp
index cba3dd919dafb..98ab86bf72eb6 100644
--- a/src/hotspot/cpu/riscv/assembler_riscv.hpp
+++ b/src/hotspot/cpu/riscv/assembler_riscv.hpp
@@ -1267,6 +1267,7 @@ enum VectorMask {
   INSN(viota_m,   0b1010111, 0b010, 0b10000, 0b010100);
 
   // Vector Single-Width Floating-Point/Integer Type-Convert Instructions
+  INSN(vfcvt_x_f_v,      0b1010111, 0b001, 0b00001, 0b010010);
   INSN(vfcvt_f_x_v,      0b1010111, 0b001, 0b00011, 0b010010);
   INSN(vfcvt_rtz_x_f_v,  0b1010111, 0b001, 0b00111, 0b010010);
 
diff --git a/src/hotspot/cpu/riscv/c1_CodeStubs_riscv.cpp b/src/hotspot/cpu/riscv/c1_CodeStubs_riscv.cpp
index b7e1b7863efdb..fb81082072610 100644
--- a/src/hotspot/cpu/riscv/c1_CodeStubs_riscv.cpp
+++ b/src/hotspot/cpu/riscv/c1_CodeStubs_riscv.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999, 2023, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 1999, 2024, Oracle and/or its affiliates. All rights reserved.
  * Copyright (c) 2014, Red Hat Inc. All rights reserved.
  * Copyright (c) 2020, 2023, Huawei Technologies Co., Ltd. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
@@ -62,7 +62,7 @@ void CounterOverflowStub::emit_code(LIR_Assembler* ce) {
   __ mov_metadata(t0, m);
   ce->store_parameter(t0, 1);
   ce->store_parameter(_bci, 0);
-  __ far_call(RuntimeAddress(Runtime1::entry_for(Runtime1::counter_overflow_id)));
+  __ far_call(RuntimeAddress(Runtime1::entry_for(C1StubId::counter_overflow_id)));
   ce->add_call_info_here(_info);
   ce->verify_oop_map(_info);
   __ j(_continuation);
@@ -71,7 +71,7 @@ void CounterOverflowStub::emit_code(LIR_Assembler* ce) {
 void RangeCheckStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
   if (_info->deoptimize_on_exception()) {
-    address a = Runtime1::entry_for(Runtime1::predicate_failed_trap_id);
+    address a = Runtime1::entry_for(C1StubId::predicate_failed_trap_id);
     __ far_call(RuntimeAddress(a));
     ce->add_call_info_here(_info);
     ce->verify_oop_map(_info);
@@ -84,13 +84,13 @@ void RangeCheckStub::emit_code(LIR_Assembler* ce) {
   } else {
     __ mv(t0, _index->as_jint());
   }
-  Runtime1::StubID stub_id;
+  C1StubId stub_id;
   if (_throw_index_out_of_bounds_exception) {
-    stub_id = Runtime1::throw_index_exception_id;
+    stub_id = C1StubId::throw_index_exception_id;
   } else {
     assert(_array != LIR_Opr::nullOpr(), "sanity");
     __ mv(t1, _array->as_pointer_register());
-    stub_id = Runtime1::throw_range_check_failed_id;
+    stub_id = C1StubId::throw_range_check_failed_id;
   }
   // t0 and t1 are used as args in generate_exception_throw，
   // so use ra as the tmp register for rt_call.
@@ -106,7 +106,7 @@ PredicateFailedStub::PredicateFailedStub(CodeEmitInfo* info) {
 
 void PredicateFailedStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
-  address a = Runtime1::entry_for(Runtime1::predicate_failed_trap_id);
+  address a = Runtime1::entry_for(C1StubId::predicate_failed_trap_id);
   __ far_call(RuntimeAddress(a));
   ce->add_call_info_here(_info);
   ce->verify_oop_map(_info);
@@ -118,7 +118,7 @@ void DivByZeroStub::emit_code(LIR_Assembler* ce) {
     ce->compilation()->implicit_exception_table()->append(_offset, __ offset());
   }
   __ bind(_entry);
-  __ far_call(RuntimeAddress(Runtime1::entry_for(Runtime1::throw_div0_exception_id)));
+  __ far_call(RuntimeAddress(Runtime1::entry_for(C1StubId::throw_div0_exception_id)));
   ce->add_call_info_here(_info);
   ce->verify_oop_map(_info);
 #ifdef ASSERT
@@ -127,14 +127,14 @@ void DivByZeroStub::emit_code(LIR_Assembler* ce) {
 }
 
 // Implementation of NewInstanceStub
-NewInstanceStub::NewInstanceStub(LIR_Opr klass_reg, LIR_Opr result, ciInstanceKlass* klass, CodeEmitInfo* info, Runtime1::StubID stub_id) {
+NewInstanceStub::NewInstanceStub(LIR_Opr klass_reg, LIR_Opr result, ciInstanceKlass* klass, CodeEmitInfo* info, C1StubId stub_id) {
   _result = result;
   _klass = klass;
   _klass_reg = klass_reg;
   _info = new CodeEmitInfo(info);
-  assert(stub_id == Runtime1::new_instance_id                 ||
-         stub_id == Runtime1::fast_new_instance_id            ||
-         stub_id == Runtime1::fast_new_instance_init_check_id,
+  assert(stub_id == C1StubId::new_instance_id                 ||
+         stub_id == C1StubId::fast_new_instance_id            ||
+         stub_id == C1StubId::fast_new_instance_init_check_id,
          "need new_instance id");
   _stub_id = stub_id;
 }
@@ -163,7 +163,7 @@ void NewTypeArrayStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
   assert(_length->as_register() == x9, "length must in x9");
   assert(_klass_reg->as_register() == x13, "klass_reg must in x13");
-  __ far_call(RuntimeAddress(Runtime1::entry_for(Runtime1::new_type_array_id)));
+  __ far_call(RuntimeAddress(Runtime1::entry_for(C1StubId::new_type_array_id)));
   ce->add_call_info_here(_info);
   ce->verify_oop_map(_info);
   assert(_result->as_register() == x10, "result must in x10");
@@ -183,7 +183,7 @@ void NewObjectArrayStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
   assert(_length->as_register() == x9, "length must in x9");
   assert(_klass_reg->as_register() == x13, "klass_reg must in x13");
-  __ far_call(RuntimeAddress(Runtime1::entry_for(Runtime1::new_object_array_id)));
+  __ far_call(RuntimeAddress(Runtime1::entry_for(C1StubId::new_object_array_id)));
   ce->add_call_info_here(_info);
   ce->verify_oop_map(_info);
   assert(_result->as_register() == x10, "result must in x10");
@@ -195,11 +195,11 @@ void MonitorEnterStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
   ce->store_parameter(_obj_reg->as_register(),  1);
   ce->store_parameter(_lock_reg->as_register(), 0);
-  Runtime1::StubID enter_id;
+  C1StubId enter_id;
   if (ce->compilation()->has_fpu_code()) {
-    enter_id = Runtime1::monitorenter_id;
+    enter_id = C1StubId::monitorenter_id;
   } else {
-    enter_id = Runtime1::monitorenter_nofpu_id;
+    enter_id = C1StubId::monitorenter_nofpu_id;
   }
   __ far_call(RuntimeAddress(Runtime1::entry_for(enter_id)));
   ce->add_call_info_here(_info);
@@ -215,11 +215,11 @@ void MonitorExitStub::emit_code(LIR_Assembler* ce) {
   }
   ce->store_parameter(_lock_reg->as_register(), 0);
   // note: non-blocking leaf routine => no call info needed
-  Runtime1::StubID exit_id;
+  C1StubId exit_id;
   if (ce->compilation()->has_fpu_code()) {
-    exit_id = Runtime1::monitorexit_id;
+    exit_id = C1StubId::monitorexit_id;
   } else {
-    exit_id = Runtime1::monitorexit_nofpu_id;
+    exit_id = C1StubId::monitorexit_nofpu_id;
   }
   __ la(ra, _continuation);
   __ far_jump(RuntimeAddress(Runtime1::entry_for(exit_id)));
@@ -244,7 +244,7 @@ void PatchingStub::emit_code(LIR_Assembler* ce) {
 void DeoptimizeStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
   ce->store_parameter(_trap_request, 0);
-  __ far_call(RuntimeAddress(Runtime1::entry_for(Runtime1::deoptimize_id)));
+  __ far_call(RuntimeAddress(Runtime1::entry_for(C1StubId::deoptimize_id)));
   ce->add_call_info_here(_info);
   DEBUG_ONLY(__ should_not_reach_here());
 }
@@ -253,9 +253,9 @@ void ImplicitNullCheckStub::emit_code(LIR_Assembler* ce) {
   address a = nullptr;
   if (_info->deoptimize_on_exception()) {
     // Deoptimize, do not throw the exception, because it is probably wrong to do it here.
-    a = Runtime1::entry_for(Runtime1::predicate_failed_trap_id);
+    a = Runtime1::entry_for(C1StubId::predicate_failed_trap_id);
   } else {
-    a = Runtime1::entry_for(Runtime1::throw_null_pointer_exception_id);
+    a = Runtime1::entry_for(C1StubId::throw_null_pointer_exception_id);
   }
 
   ce->compilation()->implicit_exception_table()->append(_offset, __ offset());
diff --git a/src/hotspot/cpu/riscv/c1_LIRAssembler_arraycopy_riscv.cpp b/src/hotspot/cpu/riscv/c1_LIRAssembler_arraycopy_riscv.cpp
index a8f260acae8ce..012932189382c 100644
--- a/src/hotspot/cpu/riscv/c1_LIRAssembler_arraycopy_riscv.cpp
+++ b/src/hotspot/cpu/riscv/c1_LIRAssembler_arraycopy_riscv.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2000, 2023, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2000, 2024, Oracle and/or its affiliates. All rights reserved.
  * Copyright (c) 2020, 2022, Huawei Technologies Co., Ltd. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
@@ -223,7 +223,7 @@ void LIR_Assembler::arraycopy_type_check(Register src, Register src_pos, Registe
     __ check_klass_subtype_fast_path(src, dst, tmp, &cont, &slow, nullptr);
 
     PUSH(src, dst);
-    __ far_call(RuntimeAddress(Runtime1::entry_for(Runtime1::slow_subtype_check_id)));
+    __ far_call(RuntimeAddress(Runtime1::entry_for(C1StubId::slow_subtype_check_id)));
     POP(src, dst);
     __ bnez(dst, cont);
 
diff --git a/src/hotspot/cpu/riscv/c1_LIRAssembler_riscv.cpp b/src/hotspot/cpu/riscv/c1_LIRAssembler_riscv.cpp
index 3d146b87707aa..940706b0a7376 100644
--- a/src/hotspot/cpu/riscv/c1_LIRAssembler_riscv.cpp
+++ b/src/hotspot/cpu/riscv/c1_LIRAssembler_riscv.cpp
@@ -305,7 +305,7 @@ int LIR_Assembler::emit_exception_handler() {
   __ verify_not_null_oop(x10);
 
   // search an exception handler (x10: exception oop, x13: throwing pc)
-  __ far_call(RuntimeAddress(Runtime1::entry_for(Runtime1::handle_exception_from_callee_id)));
+  __ far_call(RuntimeAddress(Runtime1::entry_for(C1StubId::handle_exception_from_callee_id)));
   __ should_not_reach_here();
   guarantee(code_offset() - offset <= exception_handler_size(), "overflow");
   __ end_a_stub();
@@ -361,7 +361,7 @@ int LIR_Assembler::emit_unwind_handler() {
   // remove the activation and dispatch to the unwind handler
   __ block_comment("remove_frame and dispatch to the unwind handler");
   __ remove_frame(initial_frame_size_in_bytes());
-  __ far_jump(RuntimeAddress(Runtime1::entry_for(Runtime1::unwind_exception_id)));
+  __ far_jump(RuntimeAddress(Runtime1::entry_for(C1StubId::unwind_exception_id)));
 
   // Emit the slow path assembly
   if (stub != nullptr) {
@@ -1088,7 +1088,7 @@ void LIR_Assembler::typecheck_helper_slowcheck(ciKlass *k, Register obj, Registe
       __ addi(sp, sp, -2 * wordSize); // 2: store k_RInfo and klass_RInfo
       __ sd(k_RInfo, Address(sp, 0));             // sub klass
       __ sd(klass_RInfo, Address(sp, wordSize));  // super klass
-      __ far_call(RuntimeAddress(Runtime1::entry_for(Runtime1::slow_subtype_check_id)));
+      __ far_call(RuntimeAddress(Runtime1::entry_for(C1StubId::slow_subtype_check_id)));
       // load result to k_RInfo
       __ ld(k_RInfo, Address(sp, 0));
       __ addi(sp, sp, 2 * wordSize); // 2: pop out k_RInfo and klass_RInfo
@@ -1103,7 +1103,7 @@ void LIR_Assembler::typecheck_helper_slowcheck(ciKlass *k, Register obj, Registe
     __ addi(sp, sp, -2 * wordSize); // 2: store k_RInfo and klass_RInfo
     __ sd(klass_RInfo, Address(sp, wordSize));  // sub klass
     __ sd(k_RInfo, Address(sp, 0));             // super klass
-    __ far_call(RuntimeAddress(Runtime1::entry_for(Runtime1::slow_subtype_check_id)));
+    __ far_call(RuntimeAddress(Runtime1::entry_for(C1StubId::slow_subtype_check_id)));
     // load result to k_RInfo
     __ ld(k_RInfo, Address(sp, 0));
     __ addi(sp, sp, 2 * wordSize); // 2: pop out k_RInfo and klass_RInfo
@@ -1391,7 +1391,7 @@ void LIR_Assembler::throw_op(LIR_Opr exceptionPC, LIR_Opr exceptionOop, CodeEmit
   // exception object is not added to oop map by LinearScan
   // (LinearScan assumes that no oops are in fixed registers)
   info->add_register_oop(exceptionOop);
-  Runtime1::StubID unwind_id;
+  C1StubId unwind_id;
 
   // get current pc information
   // pc is only needed if the method has an exception handler, the unwind code does not need it.
@@ -1414,9 +1414,9 @@ void LIR_Assembler::throw_op(LIR_Opr exceptionPC, LIR_Opr exceptionOop, CodeEmit
   __ verify_not_null_oop(x10);
   // search an exception handler (x10: exception oop, x13: throwing pc)
   if (compilation()->has_fpu_code()) {
-    unwind_id = Runtime1::handle_exception_id;
+    unwind_id = C1StubId::handle_exception_id;
   } else {
-    unwind_id = Runtime1::handle_exception_nofpu_id;
+    unwind_id = C1StubId::handle_exception_nofpu_id;
   }
   __ far_call(RuntimeAddress(Runtime1::entry_for(unwind_id)));
   __ nop();
@@ -2054,16 +2054,16 @@ void LIR_Assembler::deoptimize_trap(CodeEmitInfo *info) {
 
   switch (patching_id(info)) {
     case PatchingStub::access_field_id:
-      target = Runtime1::entry_for(Runtime1::access_field_patching_id);
+      target = Runtime1::entry_for(C1StubId::access_field_patching_id);
       break;
     case PatchingStub::load_klass_id:
-      target = Runtime1::entry_for(Runtime1::load_klass_patching_id);
+      target = Runtime1::entry_for(C1StubId::load_klass_patching_id);
       break;
     case PatchingStub::load_mirror_id:
-      target = Runtime1::entry_for(Runtime1::load_mirror_patching_id);
+      target = Runtime1::entry_for(C1StubId::load_mirror_patching_id);
       break;
     case PatchingStub::load_appendix_id:
-      target = Runtime1::entry_for(Runtime1::load_appendix_patching_id);
+      target = Runtime1::entry_for(C1StubId::load_appendix_patching_id);
       break;
     default: ShouldNotReachHere();
   }
@@ -2152,7 +2152,7 @@ void LIR_Assembler::lir_store_slowcheck(Register k_RInfo, Register klass_RInfo,
   __ addi(sp, sp, -2 * wordSize); // 2: store k_RInfo and klass_RInfo
   __ sd(klass_RInfo, Address(sp, wordSize));  // sub klass
   __ sd(k_RInfo, Address(sp, 0));             // super klass
-  __ far_call(RuntimeAddress(Runtime1::entry_for(Runtime1::slow_subtype_check_id)));
+  __ far_call(RuntimeAddress(Runtime1::entry_for(C1StubId::slow_subtype_check_id)));
   // load result to k_RInfo
   __ ld(k_RInfo, Address(sp, 0));
   __ addi(sp, sp, 2 * wordSize); // 2: pop out k_RInfo and klass_RInfo
diff --git a/src/hotspot/cpu/riscv/c1_LIRGenerator_riscv.cpp b/src/hotspot/cpu/riscv/c1_LIRGenerator_riscv.cpp
index 409e8dc0a0d95..b328d457192ba 100644
--- a/src/hotspot/cpu/riscv/c1_LIRGenerator_riscv.cpp
+++ b/src/hotspot/cpu/riscv/c1_LIRGenerator_riscv.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2005, 2023, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2005, 2024, Oracle and/or its affiliates. All rights reserved.
  * Copyright (c) 2014, Red Hat Inc. All rights reserved.
  * Copyright (c) 2020, 2022, Huawei Technologies Co., Ltd. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
@@ -1030,7 +1030,7 @@ void LIRGenerator::do_NewMultiArray(NewMultiArray* x) {
   args->append(rank);
   args->append(varargs);
   LIR_Opr reg = result_register_for(x->type());
-  __ call_runtime(Runtime1::entry_for(Runtime1::new_multi_array_id),
+  __ call_runtime(Runtime1::entry_for(C1StubId::new_multi_array_id),
                   LIR_OprFact::illegalOpr,
                   reg, args, info);
 
@@ -1062,7 +1062,7 @@ void LIRGenerator::do_CheckCast(CheckCast* x) {
   CodeStub* stub = nullptr;
   if (x->is_incompatible_class_change_check()) {
     assert(patching_info == nullptr, "can't patch this");
-    stub = new SimpleExceptionStub(Runtime1::throw_incompatible_class_change_error_id, LIR_OprFact::illegalOpr,
+    stub = new SimpleExceptionStub(C1StubId::throw_incompatible_class_change_error_id, LIR_OprFact::illegalOpr,
                                    info_for_exception);
   } else if (x->is_invokespecial_receiver_check()) {
     assert(patching_info == nullptr, "can't patch this");
@@ -1070,7 +1070,7 @@ void LIRGenerator::do_CheckCast(CheckCast* x) {
                               Deoptimization::Reason_class_check,
                               Deoptimization::Action_none);
   } else {
-    stub = new SimpleExceptionStub(Runtime1::throw_class_cast_exception_id, obj.result(), info_for_exception);
+    stub = new SimpleExceptionStub(C1StubId::throw_class_cast_exception_id, obj.result(), info_for_exception);
   }
   LIR_Opr reg = rlock_result(x);
   LIR_Opr tmp3 = LIR_OprFact::illegalOpr;
diff --git a/src/hotspot/cpu/riscv/c1_MacroAssembler_riscv.cpp b/src/hotspot/cpu/riscv/c1_MacroAssembler_riscv.cpp
index 1ae64b4f283ba..1e4b66069ee23 100644
--- a/src/hotspot/cpu/riscv/c1_MacroAssembler_riscv.cpp
+++ b/src/hotspot/cpu/riscv/c1_MacroAssembler_riscv.cpp
@@ -276,7 +276,7 @@ void C1_MacroAssembler::initialize_object(Register obj, Register klass, Register
 
   if (CURRENT_ENV->dtrace_alloc_probes()) {
     assert(obj == x10, "must be");
-    far_call(RuntimeAddress(Runtime1::entry_for(Runtime1::dtrace_object_alloc_id)));
+    far_call(RuntimeAddress(Runtime1::entry_for(C1StubId::dtrace_object_alloc_id)));
   }
 
   verify_oop(obj);
@@ -316,7 +316,7 @@ void C1_MacroAssembler::allocate_array(Register obj, Register len, Register tmp1
 
   if (CURRENT_ENV->dtrace_alloc_probes()) {
     assert(obj == x10, "must be");
-    far_call(RuntimeAddress(Runtime1::entry_for(Runtime1::dtrace_object_alloc_id)));
+    far_call(RuntimeAddress(Runtime1::entry_for(C1StubId::dtrace_object_alloc_id)));
   }
 
   verify_oop(obj);
diff --git a/src/hotspot/cpu/riscv/c1_Runtime1_riscv.cpp b/src/hotspot/cpu/riscv/c1_Runtime1_riscv.cpp
index 824d03640517e..5e4031727c827 100644
--- a/src/hotspot/cpu/riscv/c1_Runtime1_riscv.cpp
+++ b/src/hotspot/cpu/riscv/c1_Runtime1_riscv.cpp
@@ -98,10 +98,10 @@ int StubAssembler::call_RT(Register oop_result, Register metadata_result, addres
     if (frame_size() == no_frame_size) {
       leave();
       far_jump(RuntimeAddress(StubRoutines::forward_exception_entry()));
-    } else if (_stub_id == Runtime1::forward_exception_id) {
+    } else if (_stub_id == (int)C1StubId::forward_exception_id) {
       should_not_reach_here();
     } else {
-      far_jump(RuntimeAddress(Runtime1::entry_for(Runtime1::forward_exception_id)));
+      far_jump(RuntimeAddress(Runtime1::entry_for(C1StubId::forward_exception_id)));
     }
     bind(L);
   }
@@ -376,7 +376,7 @@ OopMapSet* Runtime1::generate_exception_throw(StubAssembler* sasm, address targe
   return oop_maps;
 }
 
-OopMapSet* Runtime1::generate_handle_exception(StubID id, StubAssembler *sasm) {
+OopMapSet* Runtime1::generate_handle_exception(C1StubId id, StubAssembler *sasm) {
   __ block_comment("generate_handle_exception");
 
   // incoming parameters
@@ -388,7 +388,7 @@ OopMapSet* Runtime1::generate_handle_exception(StubID id, StubAssembler *sasm) {
   OopMap* oop_map = nullptr;
 
   switch (id) {
-    case forward_exception_id:
+    case C1StubId::forward_exception_id:
       // We're handling an exception in the context of a compiled frame.
       // The registers have been saved in the standard places.  Perform
       // an exception lookup in the caller and dispatch to the handler
@@ -407,12 +407,12 @@ OopMapSet* Runtime1::generate_handle_exception(StubID id, StubAssembler *sasm) {
       __ sd(zr, Address(xthread, JavaThread::vm_result_offset()));
       __ sd(zr, Address(xthread, JavaThread::vm_result_2_offset()));
       break;
-    case handle_exception_nofpu_id:
-    case handle_exception_id:
+    case C1StubId::handle_exception_nofpu_id:
+    case C1StubId::handle_exception_id:
       // At this point all registers MAY be live.
-      oop_map = save_live_registers(sasm, id != handle_exception_nofpu_id);
+      oop_map = save_live_registers(sasm, id != C1StubId::handle_exception_nofpu_id);
       break;
-    case handle_exception_from_callee_id: {
+    case C1StubId::handle_exception_from_callee_id: {
       // At this point all registers except exception oop (x10) and
       // exception pc (ra) are dead.
       const int frame_size = 2 /* fp, return address */;
@@ -469,13 +469,13 @@ OopMapSet* Runtime1::generate_handle_exception(StubID id, StubAssembler *sasm) {
   __ sd(x10, Address(fp, frame::return_addr_offset * BytesPerWord));
 
   switch (id) {
-    case forward_exception_id:
-    case handle_exception_nofpu_id:
-    case handle_exception_id:
+    case C1StubId::forward_exception_id:
+    case C1StubId::handle_exception_nofpu_id:
+    case C1StubId::handle_exception_id:
       // Restore the registers that were saved at the beginning.
-      restore_live_registers(sasm, id != handle_exception_nofpu_id);
+      restore_live_registers(sasm, id != C1StubId::handle_exception_nofpu_id);
       break;
-    case handle_exception_from_callee_id:
+    case C1StubId::handle_exception_from_callee_id:
       break;
     default: ShouldNotReachHere();
   }
@@ -621,7 +621,7 @@ OopMapSet* Runtime1::generate_patching(StubAssembler* sasm, address target) {
   return oop_maps;
 }
 
-OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
+OopMapSet* Runtime1::generate_code_for(C1StubId id, StubAssembler* sasm) {
   // for better readability
   const bool dont_gc_arguments = false;
 
@@ -632,7 +632,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
   OopMapSet* oop_maps = nullptr;
   switch (id) {
     {
-    case forward_exception_id:
+    case C1StubId::forward_exception_id:
       {
         oop_maps = generate_handle_exception(id, sasm);
         __ leave();
@@ -640,32 +640,32 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case throw_div0_exception_id:
+    case C1StubId::throw_div0_exception_id:
       {
         StubFrame f(sasm, "throw_div0_exception", dont_gc_arguments, does_not_return);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_div0_exception), false);
       }
       break;
 
-    case throw_null_pointer_exception_id:
+    case C1StubId::throw_null_pointer_exception_id:
       { StubFrame f(sasm, "throw_null_pointer_exception", dont_gc_arguments, does_not_return);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_null_pointer_exception), false);
       }
       break;
 
-    case new_instance_id:
-    case fast_new_instance_id:
-    case fast_new_instance_init_check_id:
+    case C1StubId::new_instance_id:
+    case C1StubId::fast_new_instance_id:
+    case C1StubId::fast_new_instance_init_check_id:
       {
         Register klass = x13; // Incoming
         Register obj   = x10; // Result
 
-        if (id == new_instance_id) {
+        if (id == C1StubId::new_instance_id) {
           __ set_info("new_instance", dont_gc_arguments);
-        } else if (id == fast_new_instance_id) {
+        } else if (id == C1StubId::fast_new_instance_id) {
           __ set_info("fast new_instance", dont_gc_arguments);
         } else {
-          assert(id == fast_new_instance_init_check_id, "bad StubID");
+          assert(id == C1StubId::fast_new_instance_init_check_id, "bad C1StubId");
           __ set_info("fast new_instance init check", dont_gc_arguments);
         }
 
@@ -686,7 +686,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
 
       break;
 
-    case counter_overflow_id:
+    case C1StubId::counter_overflow_id:
       {
         Register bci = x10;
         Register method = x11;
@@ -710,14 +710,14 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case new_type_array_id:
-    case new_object_array_id:
+    case C1StubId::new_type_array_id:
+    case C1StubId::new_object_array_id:
       {
         Register length   = x9;  // Incoming
         Register klass    = x13; // Incoming
         Register obj      = x10; // Result
 
-        if (id == new_type_array_id) {
+        if (id == C1StubId::new_type_array_id) {
           __ set_info("new_type_array", dont_gc_arguments);
         } else {
           __ set_info("new_object_array", dont_gc_arguments);
@@ -730,7 +730,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
           Register tmp = obj;
           __ lwu(tmp, Address(klass, Klass::layout_helper_offset()));
           __ sraiw(tmp, tmp, Klass::_lh_array_tag_shift);
-          int tag = ((id == new_type_array_id) ? Klass::_lh_array_tag_type_value : Klass::_lh_array_tag_obj_value);
+          int tag = ((id == C1StubId::new_type_array_id) ? Klass::_lh_array_tag_type_value : Klass::_lh_array_tag_obj_value);
           __ mv(t0, tag);
           __ beq(t0, tmp, ok);
           __ stop("assert(is an array klass)");
@@ -743,7 +743,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
         OopMap* map = save_live_registers(sasm);
         assert_cond(map != nullptr);
         int call_offset = 0;
-        if (id == new_type_array_id) {
+        if (id == C1StubId::new_type_array_id) {
           call_offset = __ call_RT(obj, noreg, CAST_FROM_FN_PTR(address, new_type_array), klass, length);
         } else {
           call_offset = __ call_RT(obj, noreg, CAST_FROM_FN_PTR(address, new_object_array), klass, length);
@@ -762,7 +762,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case new_multi_array_id:
+    case C1StubId::new_multi_array_id:
       {
         StubFrame f(sasm, "new_multi_array", dont_gc_arguments);
         // x10: klass
@@ -785,7 +785,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case register_finalizer_id:
+    case C1StubId::register_finalizer_id:
       {
         __ set_info("register_finalizer", dont_gc_arguments);
 
@@ -819,14 +819,14 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case throw_class_cast_exception_id:
+    case C1StubId::throw_class_cast_exception_id:
       {
         StubFrame f(sasm, "throw_class_cast_exception", dont_gc_arguments, does_not_return);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_class_cast_exception), true);
       }
       break;
 
-    case throw_incompatible_class_change_error_id:
+    case C1StubId::throw_incompatible_class_change_error_id:
       {
         StubFrame f(sasm, "throw_incompatible_class_cast_exception", dont_gc_arguments, does_not_return);
         oop_maps = generate_exception_throw(sasm,
@@ -834,7 +834,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case slow_subtype_check_id:
+    case C1StubId::slow_subtype_check_id:
       {
         // Typical calling sequence:
         // push klass_RInfo (object klass or other subclass)
@@ -874,10 +874,10 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case monitorenter_nofpu_id:
+    case C1StubId::monitorenter_nofpu_id:
       save_fpu_registers = false;
       // fall through
-    case monitorenter_id:
+    case C1StubId::monitorenter_id:
       {
         StubFrame f(sasm, "monitorenter", dont_gc_arguments);
         OopMap* map = save_live_registers(sasm, save_fpu_registers);
@@ -896,10 +896,10 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case monitorexit_nofpu_id:
+    case C1StubId::monitorexit_nofpu_id:
       save_fpu_registers = false;
       // fall through
-    case monitorexit_id:
+    case C1StubId::monitorexit_id:
       {
         StubFrame f(sasm, "monitorexit", dont_gc_arguments);
         OopMap* map = save_live_registers(sasm, save_fpu_registers);
@@ -920,7 +920,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case deoptimize_id:
+    case C1StubId::deoptimize_id:
       {
         StubFrame f(sasm, "deoptimize", dont_gc_arguments, does_not_return);
         OopMap* oop_map = save_live_registers(sasm);
@@ -939,14 +939,14 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case throw_range_check_failed_id:
+    case C1StubId::throw_range_check_failed_id:
       {
         StubFrame f(sasm, "range_check_failed", dont_gc_arguments, does_not_return);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_range_check_exception), true);
       }
       break;
 
-    case unwind_exception_id:
+    case C1StubId::unwind_exception_id:
       {
         __ set_info("unwind_exception", dont_gc_arguments);
         // note: no stubframe since we are about to leave the current
@@ -955,7 +955,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case access_field_patching_id:
+    case C1StubId::access_field_patching_id:
       {
         StubFrame f(sasm, "access_field_patching", dont_gc_arguments, does_not_return);
         // we should set up register map
@@ -963,7 +963,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case load_klass_patching_id:
+    case C1StubId::load_klass_patching_id:
       {
         StubFrame f(sasm, "load_klass_patching", dont_gc_arguments, does_not_return);
         // we should set up register map
@@ -971,7 +971,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case load_mirror_patching_id:
+    case C1StubId::load_mirror_patching_id:
       {
         StubFrame f(sasm, "load_mirror_patching", dont_gc_arguments, does_not_return);
         // we should set up register map
@@ -979,7 +979,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case load_appendix_patching_id:
+    case C1StubId::load_appendix_patching_id:
       {
         StubFrame f(sasm, "load_appendix_patching", dont_gc_arguments, does_not_return);
         // we should set up register map
@@ -987,29 +987,29 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case handle_exception_nofpu_id:
-    case handle_exception_id:
+    case C1StubId::handle_exception_nofpu_id:
+    case C1StubId::handle_exception_id:
       {
         StubFrame f(sasm, "handle_exception", dont_gc_arguments);
         oop_maps = generate_handle_exception(id, sasm);
       }
       break;
 
-    case handle_exception_from_callee_id:
+    case C1StubId::handle_exception_from_callee_id:
       {
         StubFrame f(sasm, "handle_exception_from_callee", dont_gc_arguments);
         oop_maps = generate_handle_exception(id, sasm);
       }
       break;
 
-    case throw_index_exception_id:
+    case C1StubId::throw_index_exception_id:
       {
         StubFrame f(sasm, "index_range_check_failed", dont_gc_arguments, does_not_return);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_index_exception), true);
       }
       break;
 
-    case throw_array_store_exception_id:
+    case C1StubId::throw_array_store_exception_id:
       {
         StubFrame f(sasm, "throw_array_store_exception", dont_gc_arguments, does_not_return);
         // tos + 0: link
@@ -1018,7 +1018,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case predicate_failed_trap_id:
+    case C1StubId::predicate_failed_trap_id:
       {
         StubFrame f(sasm, "predicate_failed_trap", dont_gc_arguments, does_not_return);
 
@@ -1038,7 +1038,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case dtrace_object_alloc_id:
+    case C1StubId::dtrace_object_alloc_id:
       { // c_rarg0: object
         StubFrame f(sasm, "dtrace_object_alloc", dont_gc_arguments);
         save_live_registers(sasm);
diff --git a/src/hotspot/cpu/riscv/c2_MacroAssembler_riscv.cpp b/src/hotspot/cpu/riscv/c2_MacroAssembler_riscv.cpp
index 1e3a8bde064b3..e2c9b9dd609e0 100644
--- a/src/hotspot/cpu/riscv/c2_MacroAssembler_riscv.cpp
+++ b/src/hotspot/cpu/riscv/c2_MacroAssembler_riscv.cpp
@@ -2385,6 +2385,74 @@ void C2_MacroAssembler::expand_bits_l_v(Register dst, Register src, Register mas
   expand_bits_v(dst, src, mask, /* is_long */ true);
 }
 
+// j.l.Math.round(float)
+//  Returns the closest int to the argument, with ties rounding to positive infinity.
+// We need to handle 3 special cases defined by java api spec:
+//    NaN,
+//    float >= Integer.MAX_VALUE,
+//    float <= Integer.MIN_VALUE.
+void C2_MacroAssembler::java_round_float_v(VectorRegister dst, VectorRegister src, FloatRegister ftmp,
+                                           BasicType bt, uint vector_length) {
+  // In riscv, there is no straight corresponding rounding mode to satisfy the behaviour defined,
+  // in java api spec, i.e. any rounding mode can not handle some corner cases, e.g.
+  //  RNE is the closest one, but it ties to "even", which means 1.5/2.5 both will be converted
+  //    to 2, instead of 2 and 3 respectively.
+  //  RUP does not work either, although java api requires "rounding to positive infinity",
+  //    but both 1.3/1.8 will be converted to 2, instead of 1 and 2 respectively.
+  //
+  // The optimal solution for non-NaN cases is:
+  //    src+0.5 => dst, with rdn rounding mode,
+  //    convert dst from float to int, with rnd rounding mode.
+  // and, this solution works as expected for float >= Integer.MAX_VALUE and float <= Integer.MIN_VALUE.
+  //
+  // But, we still need to handle NaN explicilty with vector mask instructions.
+  //
+  // Check MacroAssembler::java_round_float and C2_MacroAssembler::vector_round_sve in aarch64 for more details.
+
+  csrwi(CSR_FRM, C2_MacroAssembler::rdn);
+  vsetvli_helper(bt, vector_length);
+
+  // don't rearrage the instructions sequence order without performance testing.
+  // check MacroAssembler::java_round_float in riscv64 for more details.
+  mv(t0, jint_cast(0.5f));
+  fmv_w_x(ftmp, t0);
+
+  // replacing vfclass with feq as performance optimization
+  vmfeq_vv(v0, src, src);
+  // set dst = 0 in cases of NaN
+  vmv_v_x(dst, zr);
+
+  // dst = (src + 0.5) rounded down towards negative infinity
+  vfadd_vf(dst, src, ftmp, Assembler::v0_t);
+  vfcvt_x_f_v(dst, dst, Assembler::v0_t); // in RoundingMode::rdn
+
+  csrwi(CSR_FRM, C2_MacroAssembler::rne);
+}
+
+// java.lang.Math.round(double a)
+// Returns the closest long to the argument, with ties rounding to positive infinity.
+void C2_MacroAssembler::java_round_double_v(VectorRegister dst, VectorRegister src, FloatRegister ftmp,
+                                            BasicType bt, uint vector_length) {
+  // check C2_MacroAssembler::java_round_float_v above for more details.
+
+  csrwi(CSR_FRM, C2_MacroAssembler::rdn);
+  vsetvli_helper(bt, vector_length);
+
+  mv(t0, julong_cast(0.5));
+  fmv_d_x(ftmp, t0);
+
+  // replacing vfclass with feq as performance optimization
+  vmfeq_vv(v0, src, src);
+  // set dst = 0 in cases of NaN
+  vmv_v_x(dst, zr);
+
+  // dst = (src + 0.5) rounded down towards negative infinity
+  vfadd_vf(dst, src, ftmp, Assembler::v0_t);
+  vfcvt_x_f_v(dst, dst, Assembler::v0_t); // in RoundingMode::rdn
+
+  csrwi(CSR_FRM, C2_MacroAssembler::rne);
+}
+
 void C2_MacroAssembler::element_compare(Register a1, Register a2, Register result, Register cnt, Register tmp1, Register tmp2,
                                         VectorRegister vr1, VectorRegister vr2, VectorRegister vrs, bool islatin, Label &DONE,
                                         Assembler::LMUL lmul) {
diff --git a/src/hotspot/cpu/riscv/c2_MacroAssembler_riscv.hpp b/src/hotspot/cpu/riscv/c2_MacroAssembler_riscv.hpp
index 4d7f756923c24..38351565cc626 100644
--- a/src/hotspot/cpu/riscv/c2_MacroAssembler_riscv.hpp
+++ b/src/hotspot/cpu/riscv/c2_MacroAssembler_riscv.hpp
@@ -187,6 +187,9 @@
   void expand_bits_i_v(Register dst, Register src, Register mask);
   void expand_bits_l_v(Register dst, Register src, Register mask);
 
+  void java_round_float_v(VectorRegister dst, VectorRegister src, FloatRegister ftmp, BasicType bt, uint vector_length);
+  void java_round_double_v(VectorRegister dst, VectorRegister src, FloatRegister ftmp, BasicType bt, uint vector_length);
+
   void float16_to_float_v(VectorRegister dst, VectorRegister src, uint vector_length);
   void float_to_float16_v(VectorRegister dst, VectorRegister src, VectorRegister vtmp, Register tmp, uint vector_length);
 
diff --git a/src/hotspot/cpu/riscv/riscv.ad b/src/hotspot/cpu/riscv/riscv.ad
index db010c9c6c82f..05f55fd0da7af 100644
--- a/src/hotspot/cpu/riscv/riscv.ad
+++ b/src/hotspot/cpu/riscv/riscv.ad
@@ -1920,6 +1920,18 @@ bool Matcher::match_rule_supported(int opcode) {
     case Op_EncodeISOArray:
       return UseRVV;
 
+    // Current test shows that, it brings performance gain when MaxVectorSize >= 32, but brings
+    // regression when MaxVectorSize == 16. So only enable the intrinsic when MaxVectorSize >= 32.
+    case Op_RoundVF:
+      return UseRVV && MaxVectorSize >= 32;
+
+    // For double, current test shows that even with MaxVectorSize == 32, there is still some regression.
+    // Although there is no hardware to verify it for now, from the trend of performance data on hardwares
+    // (with vlenb == 16 and 32 respectively), it's promising to bring better performance rather than
+    // regression for double when MaxVectorSize == 64+. So only enable the intrinsic when MaxVectorSize >= 64.
+    case Op_RoundVD:
+      return UseRVV && MaxVectorSize >= 64;
+
     case Op_PopCountI:
     case Op_PopCountL:
       return UsePopCountInstruction;
diff --git a/src/hotspot/cpu/riscv/riscv_v.ad b/src/hotspot/cpu/riscv/riscv_v.ad
index 1a51d7583c9dd..54947f6bf9a19 100644
--- a/src/hotspot/cpu/riscv/riscv_v.ad
+++ b/src/hotspot/cpu/riscv/riscv_v.ad
@@ -4715,6 +4715,34 @@ instruct vsignum_reg(vReg dst, vReg zero, vReg one, vRegMask_V0 v0) %{
   ins_pipe(pipe_slow);
 %}
 
+// ---------------- Round float/double Vector Operations ----------------
+
+instruct vround_f(vReg dst, vReg src, fRegF tmp, vRegMask_V0 v0) %{
+  match(Set dst (RoundVF src));
+  effect(TEMP_DEF dst, TEMP tmp, TEMP v0);
+  format %{ "java_round_float_v $dst, $src\t" %}
+  ins_encode %{
+    BasicType bt = Matcher::vector_element_basic_type(this);
+    uint vector_length = Matcher::vector_length(this);
+    __ java_round_float_v(as_VectorRegister($dst$$reg), as_VectorRegister($src$$reg),
+                          as_FloatRegister($tmp$$reg), bt, vector_length);
+  %}
+  ins_pipe(pipe_slow);
+%}
+
+instruct vround_d(vReg dst, vReg src, fRegD tmp, vRegMask_V0 v0) %{
+  match(Set dst (RoundVD src));
+  effect(TEMP_DEF dst, TEMP tmp, TEMP v0);
+  format %{ "java_round_double_v $dst, $src\t" %}
+  ins_encode %{
+    BasicType bt = Matcher::vector_element_basic_type(this);
+    uint vector_length = Matcher::vector_length(this);
+    __ java_round_double_v(as_VectorRegister($dst$$reg), as_VectorRegister($src$$reg),
+                           as_FloatRegister($tmp$$reg), bt, vector_length);
+  %}
+  ins_pipe(pipe_slow);
+%}
+
 // -------------------------------- Reverse Bytes Vector Operations ------------------------
 
 instruct vreverse_bytes_masked(vReg dst_src, vRegMask_V0 v0) %{
diff --git a/src/hotspot/cpu/s390/c1_CodeStubs_s390.cpp b/src/hotspot/cpu/s390/c1_CodeStubs_s390.cpp
index b7f1d3605681a..e01e4458e38d3 100644
--- a/src/hotspot/cpu/s390/c1_CodeStubs_s390.cpp
+++ b/src/hotspot/cpu/s390/c1_CodeStubs_s390.cpp
@@ -48,7 +48,7 @@ void C1SafepointPollStub::emit_code(LIR_Assembler* ce) {
 void RangeCheckStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
   if (_info->deoptimize_on_exception()) {
-    address a = Runtime1::entry_for (Runtime1::predicate_failed_trap_id);
+    address a = Runtime1::entry_for (C1StubId::predicate_failed_trap_id);
     ce->emit_call_c(a);
     CHECK_BAILOUT();
     ce->add_call_info_here(_info);
@@ -64,11 +64,11 @@ void RangeCheckStub::emit_code(LIR_Assembler* ce) {
     __ load_const_optimized(Z_R1_scratch, _index->as_jint());
   }
 
-  Runtime1::StubID stub_id;
+  C1StubId stub_id;
   if (_throw_index_out_of_bounds_exception) {
-    stub_id = Runtime1::throw_index_exception_id;
+    stub_id = C1StubId::throw_index_exception_id;
   } else {
-    stub_id = Runtime1::throw_range_check_failed_id;
+    stub_id = C1StubId::throw_range_check_failed_id;
     __ lgr_if_needed(Z_R0_scratch, _array->as_pointer_register());
   }
   ce->emit_call_c(Runtime1::entry_for (stub_id));
@@ -84,7 +84,7 @@ PredicateFailedStub::PredicateFailedStub(CodeEmitInfo* info) {
 
 void PredicateFailedStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
-  address a = Runtime1::entry_for (Runtime1::predicate_failed_trap_id);
+  address a = Runtime1::entry_for (C1StubId::predicate_failed_trap_id);
   ce->emit_call_c(a);
   CHECK_BAILOUT();
   ce->add_call_info_here(_info);
@@ -102,7 +102,7 @@ void CounterOverflowStub::emit_code(LIR_Assembler* ce) {
   }
   ce->store_parameter(/*_method->as_register()*/ Z_R1_scratch, 1);
   ce->store_parameter(_bci, 0);
-  ce->emit_call_c(Runtime1::entry_for (Runtime1::counter_overflow_id));
+  ce->emit_call_c(Runtime1::entry_for (C1StubId::counter_overflow_id));
   CHECK_BAILOUT();
   ce->add_call_info_here(_info);
   ce->verify_oop_map(_info);
@@ -114,7 +114,7 @@ void DivByZeroStub::emit_code(LIR_Assembler* ce) {
     ce->compilation()->implicit_exception_table()->append(_offset, __ offset());
   }
   __ bind(_entry);
-  ce->emit_call_c(Runtime1::entry_for (Runtime1::throw_div0_exception_id));
+  ce->emit_call_c(Runtime1::entry_for (C1StubId::throw_div0_exception_id));
   CHECK_BAILOUT();
   ce->add_call_info_here(_info);
   debug_only(__ should_not_reach_here());
@@ -124,9 +124,9 @@ void ImplicitNullCheckStub::emit_code(LIR_Assembler* ce) {
   address a;
   if (_info->deoptimize_on_exception()) {
     // Deoptimize, do not throw the exception, because it is probably wrong to do it here.
-    a = Runtime1::entry_for (Runtime1::predicate_failed_trap_id);
+    a = Runtime1::entry_for (C1StubId::predicate_failed_trap_id);
   } else {
-    a = Runtime1::entry_for (Runtime1::throw_null_pointer_exception_id);
+    a = Runtime1::entry_for (C1StubId::throw_null_pointer_exception_id);
   }
 
   ce->compilation()->implicit_exception_table()->append(_offset, __ offset());
@@ -151,14 +151,14 @@ void SimpleExceptionStub::emit_code(LIR_Assembler* ce) {
   debug_only(__ should_not_reach_here());
 }
 
-NewInstanceStub::NewInstanceStub(LIR_Opr klass_reg, LIR_Opr result, ciInstanceKlass* klass, CodeEmitInfo* info, Runtime1::StubID stub_id) {
+NewInstanceStub::NewInstanceStub(LIR_Opr klass_reg, LIR_Opr result, ciInstanceKlass* klass, CodeEmitInfo* info, C1StubId stub_id) {
   _result = result;
   _klass = klass;
   _klass_reg = klass_reg;
   _info = new CodeEmitInfo(info);
-  assert(stub_id == Runtime1::new_instance_id                 ||
-         stub_id == Runtime1::fast_new_instance_id            ||
-         stub_id == Runtime1::fast_new_instance_init_check_id,
+  assert(stub_id == C1StubId::new_instance_id                 ||
+         stub_id == C1StubId::fast_new_instance_id            ||
+         stub_id == C1StubId::fast_new_instance_init_check_id,
          "need new_instance id");
   _stub_id = stub_id;
 }
@@ -186,7 +186,7 @@ void NewTypeArrayStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
   assert(_klass_reg->as_register() == Z_R11, "call target expects klass in Z_R11");
   __ lgr_if_needed(Z_R13, _length->as_register());
-  address a = Runtime1::entry_for (Runtime1::new_type_array_id);
+  address a = Runtime1::entry_for (C1StubId::new_type_array_id);
   ce->emit_call_c(a);
   CHECK_BAILOUT();
   ce->add_call_info_here(_info);
@@ -206,7 +206,7 @@ void NewObjectArrayStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
   assert(_klass_reg->as_register() == Z_R11, "call target expects klass in Z_R11");
   __ lgr_if_needed(Z_R13, _length->as_register());
-  address a = Runtime1::entry_for (Runtime1::new_object_array_id);
+  address a = Runtime1::entry_for (C1StubId::new_object_array_id);
   ce->emit_call_c(a);
   CHECK_BAILOUT();
   ce->add_call_info_here(_info);
@@ -217,11 +217,11 @@ void NewObjectArrayStub::emit_code(LIR_Assembler* ce) {
 
 void MonitorEnterStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
-  Runtime1::StubID enter_id;
+  C1StubId enter_id;
   if (ce->compilation()->has_fpu_code()) {
-    enter_id = Runtime1::monitorenter_id;
+    enter_id = C1StubId::monitorenter_id;
   } else {
-    enter_id = Runtime1::monitorenter_nofpu_id;
+    enter_id = C1StubId::monitorenter_nofpu_id;
   }
   __ lgr_if_needed(Z_R1_scratch, _obj_reg->as_register());
   __ lgr_if_needed(Z_R13, _lock_reg->as_register()); // See LIRGenerator::syncTempOpr().
@@ -242,11 +242,11 @@ void MonitorExitStub::emit_code(LIR_Assembler* ce) {
     __ lgr_if_needed(Z_R1_scratch, _lock_reg->as_register());
   }
   // Note: non-blocking leaf routine => no call info needed.
-  Runtime1::StubID exit_id;
+  C1StubId exit_id;
   if (ce->compilation()->has_fpu_code()) {
-    exit_id = Runtime1::monitorexit_id;
+    exit_id = C1StubId::monitorexit_id;
   } else {
-    exit_id = Runtime1::monitorexit_nofpu_id;
+    exit_id = C1StubId::monitorexit_nofpu_id;
   }
   ce->emit_call_c(Runtime1::entry_for (exit_id));
   CHECK_BAILOUT();
@@ -378,10 +378,10 @@ void PatchingStub::emit_code(LIR_Assembler* ce) {
   address target = nullptr;
   relocInfo::relocType reloc_type = relocInfo::none;
   switch (_id) {
-    case access_field_id:  target = Runtime1::entry_for (Runtime1::access_field_patching_id); break;
-    case load_klass_id:    target = Runtime1::entry_for (Runtime1::load_klass_patching_id); reloc_type = relocInfo::metadata_type; break;
-    case load_mirror_id:   target = Runtime1::entry_for (Runtime1::load_mirror_patching_id); reloc_type = relocInfo::oop_type; break;
-    case load_appendix_id: target = Runtime1::entry_for (Runtime1::load_appendix_patching_id); reloc_type = relocInfo::oop_type; break;
+    case access_field_id:  target = Runtime1::entry_for (C1StubId::access_field_patching_id); break;
+    case load_klass_id:    target = Runtime1::entry_for (C1StubId::load_klass_patching_id); reloc_type = relocInfo::metadata_type; break;
+    case load_mirror_id:   target = Runtime1::entry_for (C1StubId::load_mirror_patching_id); reloc_type = relocInfo::oop_type; break;
+    case load_appendix_id: target = Runtime1::entry_for (C1StubId::load_appendix_patching_id); reloc_type = relocInfo::oop_type; break;
     default: ShouldNotReachHere();
   }
   __ bind(call_patch);
@@ -406,7 +406,7 @@ void PatchingStub::emit_code(LIR_Assembler* ce) {
 void DeoptimizeStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
   __ load_const_optimized(Z_R1_scratch, _trap_request); // Pass trap request in Z_R1_scratch.
-  ce->emit_call_c(Runtime1::entry_for (Runtime1::deoptimize_id));
+  ce->emit_call_c(Runtime1::entry_for (C1StubId::deoptimize_id));
   CHECK_BAILOUT();
   ce->add_call_info_here(_info);
   DEBUG_ONLY(__ should_not_reach_here());
diff --git a/src/hotspot/cpu/s390/c1_LIRAssembler_s390.cpp b/src/hotspot/cpu/s390/c1_LIRAssembler_s390.cpp
index a5e62169a9350..d288f4a893d0a 100644
--- a/src/hotspot/cpu/s390/c1_LIRAssembler_s390.cpp
+++ b/src/hotspot/cpu/s390/c1_LIRAssembler_s390.cpp
@@ -172,7 +172,7 @@ int LIR_Assembler::emit_exception_handler() {
 
   int offset = code_offset();
 
-  address a = Runtime1::entry_for (Runtime1::handle_exception_from_callee_id);
+  address a = Runtime1::entry_for (C1StubId::handle_exception_from_callee_id);
   address call_addr = emit_call_c(a);
   CHECK_BAILOUT_(-1);
   __ should_not_reach_here();
@@ -212,7 +212,7 @@ int LIR_Assembler::emit_unwind_handler() {
   // Perform needed unlocking.
   MonitorExitStub* stub = nullptr;
   if (method()->is_synchronized()) {
-    // Runtime1::monitorexit_id expects lock address in Z_R1_scratch.
+    // C1StubId::monitorexit_id expects lock address in Z_R1_scratch.
     LIR_Opr lock = FrameMap::as_opr(Z_R1_scratch);
     monitor_address(0, lock);
     stub = new MonitorExitStub(lock, true, 0);
@@ -241,7 +241,7 @@ int LIR_Assembler::emit_unwind_handler() {
   // Z_EXC_PC: exception pc
 
   // Dispatch to the unwind logic.
-  __ load_const_optimized(Z_R5, Runtime1::entry_for (Runtime1::unwind_exception_id));
+  __ load_const_optimized(Z_R5, Runtime1::entry_for (C1StubId::unwind_exception_id));
   __ z_br(Z_R5);
 
   // Emit the slow path assembly.
@@ -1910,8 +1910,8 @@ void LIR_Assembler::throw_op(LIR_Opr exceptionPC, LIR_Opr exceptionOop, CodeEmit
   // Reuse the debug info from the safepoint poll for the throw op itself.
   __ get_PC(Z_EXC_PC);
   add_call_info(__ offset(), info); // for exception handler
-  address stub = Runtime1::entry_for (compilation()->has_fpu_code() ? Runtime1::handle_exception_id
-                                                                    : Runtime1::handle_exception_nofpu_id);
+  address stub = Runtime1::entry_for (compilation()->has_fpu_code() ? C1StubId::handle_exception_id
+                                                                    : C1StubId::handle_exception_nofpu_id);
   emit_call_c(stub);
 }
 
@@ -2116,7 +2116,7 @@ void LIR_Assembler::emit_arraycopy(LIR_OpArrayCopy* op) {
 
       store_parameter(src_klass, 0); // sub
       store_parameter(dst_klass, 1); // super
-      emit_call_c(Runtime1::entry_for (Runtime1::slow_subtype_check_id));
+      emit_call_c(Runtime1::entry_for (C1StubId::slow_subtype_check_id));
       CHECK_BAILOUT2(cont, slow);
       // Sets condition code 0 for match (2 otherwise).
       __ branch_optimized(Assembler::bcondEqual, cont);
@@ -2539,7 +2539,7 @@ void LIR_Assembler::emit_typecheck_helper(LIR_OpTypeCheck *op, Label* success, L
                                      RegisterOrConstant(super_check_offset));
     if (need_slow_path) {
       // Call out-of-line instance of __ check_klass_subtype_slow_path(...):
-      address a = Runtime1::entry_for (Runtime1::slow_subtype_check_id);
+      address a = Runtime1::entry_for (C1StubId::slow_subtype_check_id);
       store_parameter(klass_RInfo, 0); // sub
       store_parameter(k_RInfo, 1);     // super
       emit_call_c(a); // Sets condition code 0 for match (2 otherwise).
@@ -2614,7 +2614,7 @@ void LIR_Assembler::emit_opTypeCheck(LIR_OpTypeCheck* op) {
     // Perform the fast part of the checking logic.
     __ check_klass_subtype_fast_path(klass_RInfo, k_RInfo, Rtmp1, success_target, failure_target, nullptr);
     // Call out-of-line instance of __ check_klass_subtype_slow_path(...):
-    address a = Runtime1::entry_for (Runtime1::slow_subtype_check_id);
+    address a = Runtime1::entry_for (C1StubId::slow_subtype_check_id);
     store_parameter(klass_RInfo, 0); // sub
     store_parameter(k_RInfo, 1);     // super
     emit_call_c(a); // Sets condition code 0 for match (2 otherwise).
diff --git a/src/hotspot/cpu/s390/c1_LIRGenerator_s390.cpp b/src/hotspot/cpu/s390/c1_LIRGenerator_s390.cpp
index 619f0f7174f01..f998e86256f56 100644
--- a/src/hotspot/cpu/s390/c1_LIRGenerator_s390.cpp
+++ b/src/hotspot/cpu/s390/c1_LIRGenerator_s390.cpp
@@ -885,7 +885,7 @@ void LIRGenerator::do_NewMultiArray(NewMultiArray* x) {
   args->append(rank);
   args->append(varargs);
   LIR_Opr reg = result_register_for (x->type());
-  __ call_runtime(Runtime1::entry_for (Runtime1::new_multi_array_id),
+  __ call_runtime(Runtime1::entry_for (C1StubId::new_multi_array_id),
                   LIR_OprFact::illegalOpr,
                   reg, args, info);
 
@@ -916,14 +916,14 @@ void LIRGenerator::do_CheckCast(CheckCast* x) {
   CodeStub* stub;
   if (x->is_incompatible_class_change_check()) {
     assert(patching_info == nullptr, "can't patch this");
-    stub = new SimpleExceptionStub(Runtime1::throw_incompatible_class_change_error_id, LIR_OprFact::illegalOpr, info_for_exception);
+    stub = new SimpleExceptionStub(C1StubId::throw_incompatible_class_change_error_id, LIR_OprFact::illegalOpr, info_for_exception);
   } else if (x->is_invokespecial_receiver_check()) {
     assert(patching_info == nullptr, "can't patch this");
     stub = new DeoptimizeStub(info_for_exception,
                               Deoptimization::Reason_class_check,
                               Deoptimization::Action_none);
   } else {
-    stub = new SimpleExceptionStub(Runtime1::throw_class_cast_exception_id, obj.result(), info_for_exception);
+    stub = new SimpleExceptionStub(C1StubId::throw_class_cast_exception_id, obj.result(), info_for_exception);
   }
   LIR_Opr reg = rlock_result(x);
   LIR_Opr tmp1 = new_register(objectType);
diff --git a/src/hotspot/cpu/s390/c1_MacroAssembler_s390.cpp b/src/hotspot/cpu/s390/c1_MacroAssembler_s390.cpp
index a9140a7925ebd..f3fa19ddb31e0 100644
--- a/src/hotspot/cpu/s390/c1_MacroAssembler_s390.cpp
+++ b/src/hotspot/cpu/s390/c1_MacroAssembler_s390.cpp
@@ -254,7 +254,7 @@ void C1_MacroAssembler::initialize_object(
   // Dtrace support is unimplemented.
   //  if (CURRENT_ENV->dtrace_alloc_probes()) {
   //    assert(obj == rax, "must be");
-  //    call(RuntimeAddress(Runtime1::entry_for (Runtime1::dtrace_object_alloc_id)));
+  //    call(RuntimeAddress(Runtime1::entry_for (C1StubId::dtrace_object_alloc_id)));
   //  }
 
   verify_oop(obj, FILE_AND_LINE);
@@ -315,7 +315,7 @@ void C1_MacroAssembler::allocate_array(
   // Dtrace support is unimplemented.
   // if (CURRENT_ENV->dtrace_alloc_probes()) {
   //   assert(obj == rax, "must be");
-  //   call(RuntimeAddress(Runtime1::entry_for (Runtime1::dtrace_object_alloc_id)));
+  //   call(RuntimeAddress(Runtime1::entry_for (C1StubId::dtrace_object_alloc_id)));
   // }
 
   verify_oop(obj, FILE_AND_LINE);
diff --git a/src/hotspot/cpu/s390/c1_Runtime1_s390.cpp b/src/hotspot/cpu/s390/c1_Runtime1_s390.cpp
index 41c57043d8234..2f629c108c956 100644
--- a/src/hotspot/cpu/s390/c1_Runtime1_s390.cpp
+++ b/src/hotspot/cpu/s390/c1_Runtime1_s390.cpp
@@ -98,10 +98,10 @@ int StubAssembler::call_RT(Register oop_result1, Register metadata_result, addre
       restore_return_pc();
       load_const_optimized(Z_R1, StubRoutines::forward_exception_entry());
       z_br(Z_R1);
-    } else if (_stub_id == Runtime1::forward_exception_id) {
+    } else if (_stub_id == (int)C1StubId::forward_exception_id) {
       should_not_reach_here();
     } else {
-      load_const_optimized(Z_R1, Runtime1::entry_for (Runtime1::forward_exception_id));
+      load_const_optimized(Z_R1, Runtime1::entry_for (C1StubId::forward_exception_id));
       z_br(Z_R1);
     }
 
@@ -305,7 +305,7 @@ OopMapSet* Runtime1::generate_patching(StubAssembler* sasm, address target) {
   return oop_maps;
 }
 
-OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
+OopMapSet* Runtime1::generate_code_for(C1StubId id, StubAssembler* sasm) {
 
   // for better readability
   const bool must_gc_arguments = true;
@@ -318,26 +318,26 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
   // Stub code and info for the different stubs.
   OopMapSet* oop_maps = nullptr;
   switch (id) {
-    case forward_exception_id:
+    case C1StubId::forward_exception_id:
       {
         oop_maps = generate_handle_exception(id, sasm);
         // will not return
       }
       break;
 
-    case new_instance_id:
-    case fast_new_instance_id:
-    case fast_new_instance_init_check_id:
+    case C1StubId::new_instance_id:
+    case C1StubId::fast_new_instance_id:
+    case C1StubId::fast_new_instance_init_check_id:
       {
         Register klass    = Z_R11; // Incoming
         Register obj      = Z_R2;  // Result
 
-        if (id == new_instance_id) {
+        if (id == C1StubId::new_instance_id) {
           __ set_info("new_instance", dont_gc_arguments);
-        } else if (id == fast_new_instance_id) {
+        } else if (id == C1StubId::fast_new_instance_id) {
           __ set_info("fast new_instance", dont_gc_arguments);
         } else {
-          assert(id == fast_new_instance_init_check_id, "bad StubID");
+          assert(id == C1StubId::fast_new_instance_init_check_id, "bad C1StubId");
           __ set_info("fast new_instance init check", dont_gc_arguments);
         }
 
@@ -352,7 +352,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case counter_overflow_id:
+    case C1StubId::counter_overflow_id:
       {
         // Arguments :
         //   bci    : stack param 0
@@ -371,14 +371,14 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
         __ z_br(Z_R14);
       }
       break;
-    case new_type_array_id:
-    case new_object_array_id:
+    case C1StubId::new_type_array_id:
+    case C1StubId::new_object_array_id:
       {
         Register length   = Z_R13; // Incoming
         Register klass    = Z_R11; // Incoming
         Register obj      = Z_R2;  // Result
 
-        if (id == new_type_array_id) {
+        if (id == C1StubId::new_type_array_id) {
           __ set_info("new_type_array", dont_gc_arguments);
         } else {
           __ set_info("new_object_array", dont_gc_arguments);
@@ -391,7 +391,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
           Register t0 = obj;
           __ mem2reg_opt(t0, Address(klass, Klass::layout_helper_offset()), false);
           __ z_sra(t0, Klass::_lh_array_tag_shift);
-          int tag = ((id == new_type_array_id)
+          int tag = ((id == C1StubId::new_type_array_id)
                      ? Klass::_lh_array_tag_type_value
                      : Klass::_lh_array_tag_obj_value);
           __ compare32_and_branch(t0, tag, Assembler::bcondEqual, ok);
@@ -403,7 +403,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
 
         OopMap* map = save_live_registers_except_r2(sasm);
         int call_offset;
-        if (id == new_type_array_id) {
+        if (id == C1StubId::new_type_array_id) {
           call_offset = __ call_RT(obj, noreg, CAST_FROM_FN_PTR(address, new_type_array), klass, length);
         } else {
           call_offset = __ call_RT(obj, noreg, CAST_FROM_FN_PTR(address, new_object_array), klass, length);
@@ -418,7 +418,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case new_multi_array_id:
+    case C1StubId::new_multi_array_id:
       { __ set_info("new_multi_array", dont_gc_arguments);
         // Z_R3,: klass
         // Z_R4,: rank
@@ -436,7 +436,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case register_finalizer_id:
+    case C1StubId::register_finalizer_id:
       {
         __ set_info("register_finalizer", dont_gc_arguments);
 
@@ -459,62 +459,62 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case throw_range_check_failed_id:
+    case C1StubId::throw_range_check_failed_id:
       { __ set_info("range_check_failed", dont_gc_arguments);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_range_check_exception), true);
       }
       break;
 
-    case throw_index_exception_id:
+    case C1StubId::throw_index_exception_id:
       { __ set_info("index_range_check_failed", dont_gc_arguments);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_index_exception), true);
       }
       break;
-    case throw_div0_exception_id:
+    case C1StubId::throw_div0_exception_id:
       { __ set_info("throw_div0_exception", dont_gc_arguments);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_div0_exception), false);
       }
       break;
-    case throw_null_pointer_exception_id:
+    case C1StubId::throw_null_pointer_exception_id:
       { __ set_info("throw_null_pointer_exception", dont_gc_arguments);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_null_pointer_exception), false);
       }
       break;
-    case handle_exception_nofpu_id:
-    case handle_exception_id:
+    case C1StubId::handle_exception_nofpu_id:
+    case C1StubId::handle_exception_id:
       { __ set_info("handle_exception", dont_gc_arguments);
         oop_maps = generate_handle_exception(id, sasm);
       }
       break;
-    case handle_exception_from_callee_id:
+    case C1StubId::handle_exception_from_callee_id:
       { __ set_info("handle_exception_from_callee", dont_gc_arguments);
         oop_maps = generate_handle_exception(id, sasm);
       }
       break;
-    case unwind_exception_id:
+    case C1StubId::unwind_exception_id:
       { __ set_info("unwind_exception", dont_gc_arguments);
         // Note: no stubframe since we are about to leave the current
         // activation and we are calling a leaf VM function only.
         generate_unwind_exception(sasm);
       }
       break;
-    case throw_array_store_exception_id:
+    case C1StubId::throw_array_store_exception_id:
       { __ set_info("throw_array_store_exception", dont_gc_arguments);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_array_store_exception), true);
       }
       break;
-    case throw_class_cast_exception_id:
+    case C1StubId::throw_class_cast_exception_id:
     { // Z_R1_scratch: object
       __ set_info("throw_class_cast_exception", dont_gc_arguments);
       oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_class_cast_exception), true);
     }
     break;
-    case throw_incompatible_class_change_error_id:
+    case C1StubId::throw_incompatible_class_change_error_id:
       { __ set_info("throw_incompatible_class_cast_exception", dont_gc_arguments);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_incompatible_class_change_error), false);
       }
       break;
-    case slow_subtype_check_id:
+    case C1StubId::slow_subtype_check_id:
     {
       // Arguments :
       //   sub  : stack param 0
@@ -580,13 +580,13 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       __ z_br(Z_R14);
     }
     break;
-    case monitorenter_nofpu_id:
-    case monitorenter_id:
+    case C1StubId::monitorenter_nofpu_id:
+    case C1StubId::monitorenter_id:
       { // Z_R1_scratch : object
         // Z_R13       : lock address (see LIRGenerator::syncTempOpr())
         __ set_info("monitorenter", dont_gc_arguments);
 
-        int save_fpu_registers = (id == monitorenter_id);
+        int save_fpu_registers = (id == C1StubId::monitorenter_id);
         // Make a frame and preserve the caller's caller-save registers.
         OopMap* oop_map = save_live_registers(sasm, save_fpu_registers);
 
@@ -600,15 +600,15 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case monitorexit_nofpu_id:
-    case monitorexit_id:
+    case C1StubId::monitorexit_nofpu_id:
+    case C1StubId::monitorexit_id:
       { // Z_R1_scratch : lock address
         // Note: really a leaf routine but must setup last java sp
         //   => Use call_RT for now (speed can be improved by
         //      doing last java sp setup manually).
         __ set_info("monitorexit", dont_gc_arguments);
 
-        int save_fpu_registers = (id == monitorexit_id);
+        int save_fpu_registers = (id == C1StubId::monitorexit_id);
         // Make a frame and preserve the caller's caller-save registers.
         OopMap* oop_map = save_live_registers(sasm, save_fpu_registers);
 
@@ -622,7 +622,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case deoptimize_id:
+    case C1StubId::deoptimize_id:
       { // Args: Z_R1_scratch: trap request
         __ set_info("deoptimize", dont_gc_arguments);
         Register trap_request = Z_R1_scratch;
@@ -639,32 +639,32 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case access_field_patching_id:
+    case C1StubId::access_field_patching_id:
       { __ set_info("access_field_patching", dont_gc_arguments);
         oop_maps = generate_patching(sasm, CAST_FROM_FN_PTR(address, access_field_patching));
       }
       break;
 
-    case load_klass_patching_id:
+    case C1StubId::load_klass_patching_id:
       { __ set_info("load_klass_patching", dont_gc_arguments);
         // We should set up register map.
         oop_maps = generate_patching(sasm, CAST_FROM_FN_PTR(address, move_klass_patching));
       }
       break;
 
-    case load_mirror_patching_id:
+    case C1StubId::load_mirror_patching_id:
       { __ set_info("load_mirror_patching", dont_gc_arguments);
         oop_maps = generate_patching(sasm, CAST_FROM_FN_PTR(address, move_mirror_patching));
       }
       break;
 
-    case load_appendix_patching_id:
+    case C1StubId::load_appendix_patching_id:
       { __ set_info("load_appendix_patching", dont_gc_arguments);
         oop_maps = generate_patching(sasm, CAST_FROM_FN_PTR(address, move_appendix_patching));
       }
       break;
 #if 0
-    case dtrace_object_alloc_id:
+    case C1StubId::dtrace_object_alloc_id:
       { // rax,: object
         StubFrame f(sasm, "dtrace_object_alloc", dont_gc_arguments);
         // We can't gc here so skip the oopmap but make sure that all
@@ -679,7 +679,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case fpu2long_stub_id:
+    case C1StubId::fpu2long_stub_id:
       {
         // rax, and rdx are destroyed, but should be free since the result is returned there
         // preserve rsi,ecx
@@ -754,7 +754,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       break;
 #endif // TODO
 
-    case predicate_failed_trap_id:
+    case C1StubId::predicate_failed_trap_id:
       {
         __ set_info("predicate_failed_trap", dont_gc_arguments);
 
@@ -775,14 +775,14 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
 
     default:
       {
-        __ should_not_reach_here(FILE_AND_LINE, id);
+        __ should_not_reach_here(FILE_AND_LINE, (int)id);
       }
       break;
   }
   return oop_maps;
 }
 
-OopMapSet* Runtime1::generate_handle_exception(StubID id, StubAssembler *sasm) {
+OopMapSet* Runtime1::generate_handle_exception(C1StubId id, StubAssembler *sasm) {
   __ block_comment("generate_handle_exception");
 
   // incoming parameters: Z_EXC_OOP, Z_EXC_PC
@@ -793,7 +793,7 @@ OopMapSet* Runtime1::generate_handle_exception(StubID id, StubAssembler *sasm) {
   Register reg_fp = Z_R1_scratch;
 
   switch (id) {
-    case forward_exception_id: {
+    case C1StubId::forward_exception_id: {
       // We're handling an exception in the context of a compiled frame.
       // The registers have been saved in the standard places. Perform
       // an exception lookup in the caller and dispatch to the handler
@@ -820,13 +820,13 @@ OopMapSet* Runtime1::generate_handle_exception(StubID id, StubAssembler *sasm) {
       __ clear_mem(Address(Z_thread, JavaThread::vm_result_2_offset()), sizeof(Metadata*));
       break;
     }
-    case handle_exception_nofpu_id:
-    case handle_exception_id:
+    case C1StubId::handle_exception_nofpu_id:
+    case C1StubId::handle_exception_id:
       // At this point all registers MAY be live.
       DEBUG_ONLY(__ z_lgr(reg_fp, Z_SP);)
-      oop_map = save_live_registers(sasm, id != handle_exception_nofpu_id, Z_EXC_PC);
+      oop_map = save_live_registers(sasm, id != C1StubId::handle_exception_nofpu_id, Z_EXC_PC);
       break;
-    case handle_exception_from_callee_id: {
+    case C1StubId::handle_exception_from_callee_id: {
       // At this point all registers except Z_EXC_OOP and Z_EXC_PC are dead.
       DEBUG_ONLY(__ z_lgr(reg_fp, Z_SP);)
       __ save_return_pc(Z_EXC_PC);
@@ -875,15 +875,15 @@ OopMapSet* Runtime1::generate_handle_exception(StubID id, StubAssembler *sasm) {
   __ invalidate_registers(Z_R2);
 
   switch(id) {
-    case forward_exception_id:
-    case handle_exception_nofpu_id:
-    case handle_exception_id:
+    case C1StubId::forward_exception_id:
+    case C1StubId::handle_exception_nofpu_id:
+    case C1StubId::handle_exception_id:
       // Restore the registers that were saved at the beginning.
       __ z_lgr(Z_R1_scratch, Z_R2);   // Restoring live registers kills Z_R2.
-      restore_live_registers(sasm, id != handle_exception_nofpu_id);  // Pops as well the frame.
+      restore_live_registers(sasm, id != C1StubId::handle_exception_nofpu_id);  // Pops as well the frame.
       __ z_br(Z_R1_scratch);
       break;
-    case handle_exception_from_callee_id: {
+    case C1StubId::handle_exception_from_callee_id: {
       __ pop_frame();
       __ z_br(Z_R2); // Jump to exception handler.
     }
diff --git a/src/hotspot/cpu/x86/assembler_x86.cpp b/src/hotspot/cpu/x86/assembler_x86.cpp
index 345b779e8094e..90c2fad1f5d57 100644
--- a/src/hotspot/cpu/x86/assembler_x86.cpp
+++ b/src/hotspot/cpu/x86/assembler_x86.cpp
@@ -1385,6 +1385,14 @@ void Assembler::addl(Address dst, int32_t imm32) {
   emit_arith_operand(0x81, rax, dst, imm32);
 }
 
+void Assembler::eaddl(Register dst, Address src, int32_t imm32, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_arith_operand(0x81, rax, src, imm32);
+}
+
 void Assembler::addb(Address dst, int imm8) {
   InstructionMark im(this);
   prefix(dst);
@@ -1429,11 +1437,26 @@ void Assembler::addl(Address dst, Register src) {
   emit_operand(src, dst, 0);
 }
 
+void Assembler::eaddl(Register dst, Address src1, Register src2, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src1, dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0x01);
+  emit_operand(src2, src1, 0);
+}
+
 void Assembler::addl(Register dst, int32_t imm32) {
   prefix(dst);
   emit_arith(0x81, 0xC0, dst, imm32);
 }
 
+void Assembler::eaddl(Register dst, Register src, int32_t imm32, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  (void) evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_arith(0x81, 0xC0, src, imm32);
+}
+
 void Assembler::addl(Register dst, Address src) {
   InstructionMark im(this);
   prefix(src, dst);
@@ -1441,11 +1464,27 @@ void Assembler::addl(Register dst, Address src) {
   emit_operand(dst, src, 0);
 }
 
+void Assembler::eaddl(Register dst, Register src1, Address src2, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src2, dst->encoding(), src1->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0x03);
+  emit_operand(src1, src2, 0);
+}
+
 void Assembler::addl(Register dst, Register src) {
   (void) prefix_and_encode(dst->encoding(), src->encoding());
   emit_arith(0x03, 0xC0, dst, src);
 }
 
+void Assembler::eaddl(Register dst, Register src1, Register src2, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  (void) evex_prefix_and_encode_ndd(src1->encoding(), dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  // opcode matches gcc
+  emit_arith(0x01, 0xC0, src1, src2);
+}
+
 void Assembler::addr_nop_4() {
   assert(UseAddressNop, "no CPU support");
   // 4 bytes: NOP DWORD PTR [EAX+0]
@@ -1632,11 +1671,25 @@ void Assembler::andl(Address dst, int32_t imm32) {
   emit_arith_operand(0x81, as_Register(4), dst, imm32);
 }
 
+void Assembler::eandl(Register dst, Address src, int32_t imm32, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_arith_operand(0x81, rsp, src, imm32);
+}
+
 void Assembler::andl(Register dst, int32_t imm32) {
   prefix(dst);
   emit_arith(0x81, 0xE0, dst, imm32);
 }
 
+void Assembler::eandl(Register dst, Register src, int32_t imm32, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  (void) evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_arith(0x81, 0xE0, src, imm32);
+}
+
 void Assembler::andl(Address dst, Register src) {
   InstructionMark im(this);
   prefix(dst, src);
@@ -1651,11 +1704,27 @@ void Assembler::andl(Register dst, Address src) {
   emit_operand(dst, src, 0);
 }
 
+void Assembler::eandl(Register dst, Register src1, Address src2, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src2, dst->encoding(), src1->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0x23);
+  emit_operand(src1, src2, 0);
+}
+
 void Assembler::andl(Register dst, Register src) {
   (void) prefix_and_encode(dst->encoding(), src->encoding());
   emit_arith(0x23, 0xC0, dst, src);
 }
 
+void Assembler::eandl(Register dst, Register src1, Register src2, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  (void) evex_prefix_and_encode_ndd(src1->encoding(), dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  // opcode matches gcc
+  emit_arith(0x21, 0xC0, src1, src2);
+}
+
 void Assembler::andnl(Register dst, Register src1, Register src2) {
   assert(VM_Version::supports_bmi1(), "bit manipulation instructions not supported");
   InstructionAttr attributes(AVX_128bit, /* rex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
@@ -1803,6 +1872,12 @@ void Assembler::cmovl(Condition cc, Register dst, Register src) {
   emit_opcode_prefix_and_encoding(0x40 | cc, 0xC0, encode);
 }
 
+void Assembler::ecmovl(Condition cc, Register dst, Register src1, Register src2) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(src1->encoding(), dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes);
+  emit_int16((0x40 | cc), (0xC0 | encode));
+}
+
 void Assembler::cmovl(Condition cc, Register dst, Address src) {
   InstructionMark im(this);
   NOT_LP64(guarantee(VM_Version::supports_cmov(), "illegal instruction"));
@@ -1811,6 +1886,15 @@ void Assembler::cmovl(Condition cc, Register dst, Address src) {
   emit_operand(dst, src, 0);
 }
 
+void Assembler::ecmovl(Condition cc, Register dst, Register src1, Address src2) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src2, dst->encoding(), src1->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes);
+  emit_int8((0x40 | cc));
+  emit_operand(src1, src2, 0);
+}
+
 void Assembler::cmpb(Address dst, Register reg) {
   assert(reg->has_byte_register(), "must have byte register");
   InstructionMark im(this);
@@ -2429,6 +2513,15 @@ void Assembler::decl(Address dst) {
   emit_operand(rcx, dst, 0);
 }
 
+void Assembler::edecl(Register dst, Address src, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8((unsigned char)0xFF);
+  emit_operand(rcx, src, 0);
+}
+
 void Assembler::divsd(XMMRegister dst, Address src) {
   NOT_LP64(assert(VM_Version::supports_sse2(), ""));
   InstructionMark im(this);
@@ -2474,21 +2567,45 @@ void Assembler::idivl(Register src) {
   emit_int16((unsigned char)0xF7, (0xF8 | encode));
 }
 
+void Assembler::eidivl(Register src, bool no_flags) { // Signed
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_nf(0, 0, src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xF7, (0xF8 | encode));
+}
+
 void Assembler::divl(Register src) { // Unsigned
   int encode = prefix_and_encode(src->encoding());
   emit_int16((unsigned char)0xF7, (0xF0 | encode));
 }
 
+void Assembler::edivl(Register src, bool no_flags) { // Unsigned
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_nf(0, 0, src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xF7, (0xF0 | encode));
+}
+
 void Assembler::imull(Register src) {
   int encode = prefix_and_encode(src->encoding());
   emit_int16((unsigned char)0xF7, (0xE8 | encode));
 }
 
+void Assembler::eimull(Register src, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_nf(0, 0, src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xF7, (0xE8 | encode));
+}
+
 void Assembler::imull(Register dst, Register src) {
   int encode = prefix_and_encode(dst->encoding(), src->encoding(), true /* is_map1 */);
   emit_opcode_prefix_and_encoding((unsigned char)0xAF, 0xC0, encode);
 }
 
+void Assembler::eimull(Register dst, Register src1, Register src2, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(src1->encoding(), dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xAF, (0xC0 | encode));
+}
+
 void Assembler::imull(Register dst, Address src, int32_t value) {
   InstructionMark im(this);
   prefix(src, dst);
@@ -2503,6 +2620,22 @@ void Assembler::imull(Register dst, Address src, int32_t value) {
   }
 }
 
+void Assembler::eimull(Register dst, Address src, int32_t value, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src, 0, dst->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  if (is8bit(value)) {
+    emit_int8((unsigned char)0x6B);
+    emit_operand(dst, src, 1);
+    emit_int8(value);
+  } else {
+    emit_int8((unsigned char)0x69);
+    emit_operand(dst, src, 4);
+    emit_int32(value);
+  }
+}
+
 void Assembler::imull(Register dst, Register src, int value) {
   int encode = prefix_and_encode(dst->encoding(), src->encoding());
   if (is8bit(value)) {
@@ -2513,6 +2646,17 @@ void Assembler::imull(Register dst, Register src, int value) {
   }
 }
 
+void Assembler::eimull(Register dst, Register src, int value, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_nf(dst->encoding(), 0, src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  if (is8bit(value)) {
+    emit_int24(0x6B, (0xC0 | encode), value & 0xFF);
+  } else {
+    emit_int16(0x69, (0xC0 | encode));
+    emit_int32(value);
+  }
+}
+
 void Assembler::imull(Register dst, Address src) {
   InstructionMark im(this);
   prefix(src, dst, false, true /* is_map1 */);
@@ -2520,6 +2664,14 @@ void Assembler::imull(Register dst, Address src) {
   emit_operand(dst, src, 0);
 }
 
+void Assembler::eimull(Register dst, Register src1, Address src2, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src2, dst->encoding(), src1->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8((unsigned char)0xAF);
+  emit_operand(src1, src2, 0);
+}
 
 void Assembler::incl(Address dst) {
   // Don't use it directly. Use MacroAssembler::increment() instead.
@@ -2529,6 +2681,16 @@ void Assembler::incl(Address dst) {
   emit_operand(rax, dst, 0);
 }
 
+void Assembler::eincl(Register dst, Address src, bool no_flags) {
+  // Don't use it directly. Use MacroAssembler::increment() instead.
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8((unsigned char)0xFF);
+  emit_operand(rax, src, 0);
+}
+
 void Assembler::jcc(Condition cc, Label& L, bool maybe_short) {
   InstructionMark im(this);
   assert((0 <= cc) && (cc < 16), "illegal cc");
@@ -2696,6 +2858,13 @@ void Assembler::lzcntl(Register dst, Register src) {
   emit_opcode_prefix_and_encoding((unsigned char)0xBD, 0xC0, encode);
 }
 
+void Assembler::elzcntl(Register dst, Register src, bool no_flags) {
+  assert(VM_Version::supports_lzcnt(), "encoding is treated as BSR");
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_nf(dst->encoding(), 0, src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xF5, (0xC0 | encode));
+}
+
 void Assembler::lzcntl(Register dst, Address src) {
   assert(VM_Version::supports_lzcnt(), "encoding is treated as BSR");
   InstructionMark im(this);
@@ -2705,6 +2874,16 @@ void Assembler::lzcntl(Register dst, Address src) {
   emit_operand(dst, src, 0);
 }
 
+void Assembler::elzcntl(Register dst, Address src, bool no_flags) {
+  assert(VM_Version::supports_lzcnt(), "encoding is treated as BSR");
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_nf(src, 0, dst->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8((unsigned char)0xF5);
+  emit_operand(dst, src, 0);
+}
+
 // Emit mfence instruction
 void Assembler::mfence() {
   NOT_LP64(assert(VM_Version::supports_sse2(), "unsupported");)
@@ -3855,11 +4034,26 @@ void Assembler::mull(Address src) {
   emit_operand(rsp, src, 0);
 }
 
+void Assembler::emull(Address src, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_nf(src, 0, 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8((unsigned char)0xF7);
+  emit_operand(rsp, src, 0);
+}
+
 void Assembler::mull(Register src) {
   int encode = prefix_and_encode(src->encoding());
   emit_int16((unsigned char)0xF7, (0xE0 | encode));
 }
 
+void Assembler::emull(Register src, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_nf(0, 0, src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xF7, (0xE0 | encode));
+}
+
 void Assembler::mulsd(XMMRegister dst, Address src) {
   NOT_LP64(assert(VM_Version::supports_sse2(), ""));
   InstructionMark im(this);
@@ -3901,6 +4095,12 @@ void Assembler::negl(Register dst) {
   emit_int16((unsigned char)0xF7, (0xD8 | encode));
 }
 
+void Assembler::enegl(Register dst, Register src, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xF7, (0xD8 | encode));
+}
+
 void Assembler::negl(Address dst) {
   InstructionMark im(this);
   prefix(dst);
@@ -3908,6 +4108,15 @@ void Assembler::negl(Address dst) {
   emit_operand(as_Register(3), dst, 0);
 }
 
+void Assembler::enegl(Register dst, Address src, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8((unsigned char)0xF7);
+  emit_operand(as_Register(3), src, 0);
+}
+
 void Assembler::nop(uint i) {
 #ifdef ASSERT
   assert(i > 0, " ");
@@ -4219,17 +4428,48 @@ void Assembler::notl(Register dst) {
   emit_int16((unsigned char)0xF7, (0xD0 | encode));
 }
 
+void Assembler::enotl(Register dst, Register src) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes);
+  emit_int16((unsigned char)0xF7, (0xD0 | encode));
+}
+
+void Assembler::orw(Register dst, Register src) {
+  (void)prefix_and_encode(dst->encoding(), src->encoding());
+  emit_arith(0x0B, 0xC0, dst, src);
+}
+
+void Assembler::eorw(Register dst, Register src1, Register src2, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  (void) evex_prefix_and_encode_ndd(src1->encoding(), dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_arith(0x0B, 0xC0, src1, src2);
+}
+
 void Assembler::orl(Address dst, int32_t imm32) {
   InstructionMark im(this);
   prefix(dst);
   emit_arith_operand(0x81, rcx, dst, imm32);
 }
 
+void Assembler::eorl(Register dst, Address src, int32_t imm32, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_arith_operand(0x81, rcx, src, imm32);
+}
+
 void Assembler::orl(Register dst, int32_t imm32) {
   prefix(dst);
   emit_arith(0x81, 0xC8, dst, imm32);
 }
 
+void Assembler::eorl(Register dst, Register src, int32_t imm32, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_arith(0x81, 0xC8, src, imm32);
+}
+
 void Assembler::orl(Register dst, Address src) {
   InstructionMark im(this);
   prefix(src, dst);
@@ -4237,11 +4477,27 @@ void Assembler::orl(Register dst, Address src) {
   emit_operand(dst, src, 0);
 }
 
+void Assembler::eorl(Register dst, Register src1, Address src2, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src2, dst->encoding(), src1->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0x0B);
+  emit_operand(src1, src2, 0);
+}
+
 void Assembler::orl(Register dst, Register src) {
   (void) prefix_and_encode(dst->encoding(), src->encoding());
   emit_arith(0x0B, 0xC0, dst, src);
 }
 
+void Assembler::eorl(Register dst, Register src1, Register src2, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  (void) evex_prefix_and_encode_ndd(src1->encoding(), dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  // opcode matches gcc
+  emit_arith(0x09, 0xC0, src1, src2);
+}
+
 void Assembler::orl(Address dst, Register src) {
   InstructionMark im(this);
   prefix(dst, src);
@@ -4249,6 +4505,15 @@ void Assembler::orl(Address dst, Register src) {
   emit_operand(src, dst, 0);
 }
 
+void Assembler::eorl(Register dst, Address src1, Register src2, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src1, dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0x09);
+  emit_operand(src2, src1, 0);
+}
+
 void Assembler::orb(Address dst, int imm8) {
   InstructionMark im(this);
   prefix(dst);
@@ -4257,6 +4522,16 @@ void Assembler::orb(Address dst, int imm8) {
   emit_int8(imm8);
 }
 
+void Assembler::eorb(Register dst, Address src, int imm8, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_8bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8((unsigned char)0x80);
+  emit_operand(rcx, src, 1);
+  emit_int8(imm8);
+}
+
 void Assembler::orb(Address dst, Register src) {
   InstructionMark im(this);
   prefix(dst, src, true);
@@ -4264,6 +4539,15 @@ void Assembler::orb(Address dst, Register src) {
   emit_operand(src, dst, 0);
 }
 
+void Assembler::eorb(Register dst, Address src1, Register src2, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_8bit);
+  evex_prefix_ndd(src1, dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0x08);
+  emit_operand(src2, src1, 0);
+}
+
 void Assembler::packsswb(XMMRegister dst, XMMRegister src) {
   NOT_LP64(assert(VM_Version::supports_sse2(), ""));
   InstructionAttr attributes(AVX_128bit, /* rex_w */ false, /* legacy_mode */ _legacy_mode_bw, /* no_mask_reg */ true, /* uses_vl */ true);
@@ -5384,6 +5668,16 @@ void Assembler::popcntl(Register dst, Address src) {
   emit_operand(dst, src, 0);
 }
 
+void Assembler::epopcntl(Register dst, Address src, bool no_flags) {
+  assert(VM_Version::supports_popcnt(), "must support");
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_nf(src, 0, dst->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8((unsigned char)0x88);
+  emit_operand(dst, src, 0);
+}
+
 void Assembler::popcntl(Register dst, Register src) {
   assert(VM_Version::supports_popcnt(), "must support");
   emit_int8((unsigned char)0xF3);
@@ -5391,6 +5685,13 @@ void Assembler::popcntl(Register dst, Register src) {
   emit_opcode_prefix_and_encoding((unsigned char)0xB8, 0xC0, encode);
 }
 
+void Assembler::epopcntl(Register dst, Register src, bool no_flags) {
+  assert(VM_Version::supports_popcnt(), "must support");
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_nf(dst->encoding(), 0, src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0x88, (0xC0 | encode));
+}
+
 void Assembler::evpopcntb(XMMRegister dst, KRegister mask, XMMRegister src, bool merge, int vector_len) {
   assert(VM_Version::supports_avx512_bitalg(), "must support avx512bitalg feature");
   assert(vector_len == AVX_512bit || VM_Version::supports_avx512vl(), "");
@@ -5979,6 +6280,17 @@ void Assembler::rcll(Register dst, int imm8) {
   }
 }
 
+void Assembler::ercll(Register dst, Register src, int imm8) {
+  assert(isShiftCount(imm8), "illegal shift count");
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes);
+  if (imm8 == 1) {
+    emit_int16((unsigned char)0xD1, (0xD0 | encode));
+  } else {
+    emit_int24((unsigned char)0xC1, (0xD0 | encode), imm8);
+  }
+}
+
 void Assembler::rcpps(XMMRegister dst, XMMRegister src) {
   NOT_LP64(assert(VM_Version::supports_sse(), ""));
   InstructionAttr attributes(AVX_128bit, /* rex_w */ false, /* legacy_mode */ true, /* no_mask_reg */ true, /* uses_vl */ false);
@@ -6059,11 +6371,28 @@ void Assembler::roll(Register dst, int imm8) {
   }
 }
 
+void Assembler::eroll(Register dst, Register src, int imm8, bool no_flags) {
+  assert(isShiftCount(imm8), "illegal shift count");
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  if (imm8 == 1) {
+     emit_int16((unsigned char)0xD1, (0xC0 | encode));
+   } else {
+     emit_int24((unsigned char)0xC1, (0xc0 | encode), imm8);
+   }
+}
+
 void Assembler::roll(Register dst) {
   int encode = prefix_and_encode(dst->encoding());
   emit_int16((unsigned char)0xD3, (0xC0 | encode));
 }
 
+void Assembler::eroll(Register dst, Register src, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xD3, (0xC0 | encode));
+}
+
 void Assembler::rorl(Register dst, int imm8) {
   assert(isShiftCount(imm8), "illegal shift count");
   int encode = prefix_and_encode(dst->encoding());
@@ -6074,17 +6403,40 @@ void Assembler::rorl(Register dst, int imm8) {
   }
 }
 
+void Assembler::erorl(Register dst, Register src, int imm8, bool no_flags) {
+  assert(isShiftCount(imm8), "illegal shift count");
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  if (imm8 == 1) {
+     emit_int16((unsigned char)0xD1, (0xC8 | encode));
+   } else {
+     emit_int24((unsigned char)0xC1, (0xc8 | encode), imm8);
+   }
+}
+
 void Assembler::rorl(Register dst) {
   int encode = prefix_and_encode(dst->encoding());
   emit_int16((unsigned char)0xD3, (0xC8 | encode));
 }
 
+void Assembler::erorl(Register dst, Register src, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xD3, (0xC8 | encode));
+}
+
 #ifdef _LP64
 void Assembler::rorq(Register dst) {
   int encode = prefixq_and_encode(dst->encoding());
   emit_int16((unsigned char)0xD3, (0xC8 | encode));
 }
 
+void Assembler::erorq(Register dst, Register src, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xD3, (0xC8 | encode));
+}
+
 void Assembler::rorq(Register dst, int imm8) {
   assert(isShiftCount(imm8 >> 1), "illegal shift count");
   int encode = prefixq_and_encode(dst->encoding());
@@ -6095,11 +6447,28 @@ void Assembler::rorq(Register dst, int imm8) {
   }
 }
 
+void Assembler::erorq(Register dst, Register src, int imm8, bool no_flags) {
+  assert(isShiftCount(imm8), "illegal shift count");
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  if (imm8 == 1) {
+     emit_int16((unsigned char)0xD1, (0xC8 | encode));
+   } else {
+     emit_int24((unsigned char)0xC1, (0xC8 | encode), imm8);
+   }
+}
+
 void Assembler::rolq(Register dst) {
   int encode = prefixq_and_encode(dst->encoding());
   emit_int16((unsigned char)0xD3, (0xC0 | encode));
 }
 
+void Assembler::erolq(Register dst, Register src, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xD3, (0xC0 | encode));
+}
+
 void Assembler::rolq(Register dst, int imm8) {
   assert(isShiftCount(imm8 >> 1), "illegal shift count");
   int encode = prefixq_and_encode(dst->encoding());
@@ -6109,6 +6478,17 @@ void Assembler::rolq(Register dst, int imm8) {
     emit_int24((unsigned char)0xC1, (0xc0 | encode), imm8);
   }
 }
+
+void Assembler::erolq(Register dst, Register src, int imm8, bool no_flags) {
+  assert(isShiftCount(imm8), "illegal shift count");
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  if (imm8 == 1) {
+     emit_int16((unsigned char)0xD1, (0xC0 | encode));
+   } else {
+     emit_int24((unsigned char)0xC1, (0xc0 | encode), imm8);
+   }
+ }
 #endif
 
 void Assembler::sahf() {
@@ -6134,6 +6514,23 @@ void Assembler::sall(Address dst, int imm8) {
   }
 }
 
+void Assembler::esall(Register dst, Address src, int imm8, bool no_flags) {
+  InstructionMark im(this);
+  assert(isShiftCount(imm8), "illegal shift count");
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  if (imm8 == 1) {
+    emit_int8((unsigned char)0xD1);
+    emit_operand(as_Register(4), src, 0);
+  }
+  else {
+    emit_int8((unsigned char)0xC1);
+    emit_operand(as_Register(4), src, 1);
+    emit_int8(imm8);
+  }
+}
+
 void Assembler::sall(Address dst) {
   InstructionMark im(this);
   prefix(dst);
@@ -6141,8 +6538,17 @@ void Assembler::sall(Address dst) {
   emit_operand(as_Register(4), dst, 0);
 }
 
-void Assembler::sall(Register dst, int imm8) {
-  assert(isShiftCount(imm8), "illegal shift count");
+void Assembler::esall(Register dst, Address src, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8((unsigned char)0xD3);
+  emit_operand(as_Register(4), src, 0);
+}
+
+void Assembler::sall(Register dst, int imm8) {
+  assert(isShiftCount(imm8), "illegal shift count");
   int encode = prefix_and_encode(dst->encoding());
   if (imm8 == 1) {
     emit_int16((unsigned char)0xD1, (0xE0 | encode));
@@ -6151,11 +6557,28 @@ void Assembler::sall(Register dst, int imm8) {
   }
 }
 
+void Assembler::esall(Register dst, Register src, int imm8, bool no_flags) {
+  assert(isShiftCount(imm8), "illegal shift count");
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  if (imm8 == 1) {
+    emit_int16((unsigned char)0xD1, (0xE0 | encode));
+  } else {
+    emit_int24((unsigned char)0xC1, (0xE0 | encode), imm8);
+  }
+}
+
 void Assembler::sall(Register dst) {
   int encode = prefix_and_encode(dst->encoding());
   emit_int16((unsigned char)0xD3, (0xE0 | encode));
 }
 
+void Assembler::esall(Register dst, Register src, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xD3, (0xE0 | encode));
+}
+
 void Assembler::sarl(Address dst, int imm8) {
   assert(isShiftCount(imm8), "illegal shift count");
   InstructionMark im(this);
@@ -6171,6 +6594,23 @@ void Assembler::sarl(Address dst, int imm8) {
   }
 }
 
+void Assembler::esarl(Register dst, Address src, int imm8, bool no_flags) {
+  assert(isShiftCount(imm8), "illegal shift count");
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  if (imm8 == 1) {
+    emit_int8((unsigned char)0xD1);
+    emit_operand(as_Register(7), src, 0);
+  }
+  else {
+    emit_int8((unsigned char)0xC1);
+    emit_operand(as_Register(7), src, 1);
+    emit_int8(imm8);
+  }
+}
+
 void Assembler::sarl(Address dst) {
   InstructionMark im(this);
   prefix(dst);
@@ -6178,6 +6618,15 @@ void Assembler::sarl(Address dst) {
   emit_operand(as_Register(7), dst, 0);
 }
 
+void Assembler::esarl(Register dst, Address src, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8((unsigned char)0xD3);
+  emit_operand(as_Register(7), src, 0);
+}
+
 void Assembler::sarl(Register dst, int imm8) {
   int encode = prefix_and_encode(dst->encoding());
   assert(isShiftCount(imm8), "illegal shift count");
@@ -6188,11 +6637,28 @@ void Assembler::sarl(Register dst, int imm8) {
   }
 }
 
+void Assembler::esarl(Register dst, Register src, int imm8, bool no_flags) {
+  assert(isShiftCount(imm8), "illegal shift count");
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  if (imm8 == 1) {
+    emit_int16((unsigned char)0xD1, (0xF8 | encode));
+  } else {
+    emit_int24((unsigned char)0xC1, (0xF8 | encode), imm8);
+  }
+}
+
 void Assembler::sarl(Register dst) {
   int encode = prefix_and_encode(dst->encoding());
   emit_int16((unsigned char)0xD3, (0xF8 | encode));
 }
 
+void Assembler::esarl(Register dst, Register src, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xD3, (0xF8 | encode));
+}
+
 void Assembler::sbbl(Address dst, int32_t imm32) {
   InstructionMark im(this);
   prefix(dst);
@@ -6204,7 +6670,6 @@ void Assembler::sbbl(Register dst, int32_t imm32) {
   emit_arith(0x81, 0xD8, dst, imm32);
 }
 
-
 void Assembler::sbbl(Register dst, Address src) {
   InstructionMark im(this);
   prefix(src, dst);
@@ -6297,7 +6762,6 @@ void Assembler::sha256msg2(XMMRegister dst, XMMRegister src) {
   emit_int16((unsigned char)0xCD, (0xC0 | encode));
 }
 
-
 void Assembler::shll(Register dst, int imm8) {
   assert(isShiftCount(imm8), "illegal shift count");
   int encode = prefix_and_encode(dst->encoding());
@@ -6308,11 +6772,28 @@ void Assembler::shll(Register dst, int imm8) {
   }
 }
 
+void Assembler::eshll(Register dst, Register src, int imm8, bool no_flags) {
+  assert(isShiftCount(imm8), "illegal shift count");
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  if (imm8 == 1 ) {
+    emit_int16((unsigned char)0xD1, (0xE0 | encode));
+  } else {
+    emit_int24((unsigned char)0xC1, (0xE0 | encode), imm8);
+  }
+}
+
 void Assembler::shll(Register dst) {
   int encode = prefix_and_encode(dst->encoding());
   emit_int16((unsigned char)0xD3, (0xE0 | encode));
 }
 
+void Assembler::eshll(Register dst, Register src, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xD3, (0xE0 | encode));
+}
+
 void Assembler::shrl(Register dst, int imm8) {
   assert(isShiftCount(imm8), "illegal shift count");
   int encode = prefix_and_encode(dst->encoding());
@@ -6324,11 +6805,29 @@ void Assembler::shrl(Register dst, int imm8) {
   }
 }
 
+void Assembler::eshrl(Register dst, Register src, int imm8, bool no_flags) {
+  assert(isShiftCount(imm8), "illegal shift count");
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  if (imm8 == 1) {
+    emit_int16((unsigned char)0xD1, (0xE8 | encode));
+  }
+  else {
+    emit_int24((unsigned char)0xC1, (0xE8 | encode), imm8);
+  }
+}
+
 void Assembler::shrl(Register dst) {
   int encode = prefix_and_encode(dst->encoding());
   emit_int16((unsigned char)0xD3, (0xE8 | encode));
 }
 
+void Assembler::eshrl(Register dst, Register src, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xD3, (0xE8 | encode));
+}
+
 void Assembler::shrl(Address dst) {
   InstructionMark im(this);
   prefix(dst);
@@ -6336,6 +6835,15 @@ void Assembler::shrl(Address dst) {
   emit_operand(as_Register(5), dst, 0);
 }
 
+void Assembler::eshrl(Register dst, Address src, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8((unsigned char)0xD3);
+  emit_operand(as_Register(5), src, 0);
+}
+
 void Assembler::shrl(Address dst, int imm8) {
   InstructionMark im(this);
   assert(isShiftCount(imm8), "illegal shift count");
@@ -6351,37 +6859,89 @@ void Assembler::shrl(Address dst, int imm8) {
   }
 }
 
+void Assembler::eshrl(Register dst, Address src, int imm8, bool no_flags) {
+  InstructionMark im(this);
+  assert(isShiftCount(imm8), "illegal shift count");
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  if (imm8 == 1) {
+    emit_int8((unsigned char)0xD1);
+    emit_operand(as_Register(5), src, 0);
+  }
+  else {
+    emit_int8((unsigned char)0xC1);
+    emit_operand(as_Register(5), src, 1);
+    emit_int8(imm8);
+  }
+}
 
 void Assembler::shldl(Register dst, Register src) {
   int encode = prefix_and_encode(src->encoding(), dst->encoding(), true /* is_map1 */);
   emit_opcode_prefix_and_encoding((unsigned char)0xA5, 0xC0, encode);
 }
 
+void Assembler::eshldl(Register dst, Register src1, Register src2, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(src1->encoding(), dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16(0xA5, (0xC0 | encode));
+}
+
 void Assembler::shldl(Register dst, Register src, int8_t imm8) {
   int encode = prefix_and_encode(src->encoding(), dst->encoding(), true /* is_map1 */);
   emit_opcode_prefix_and_encoding((unsigned char)0xA4, 0xC0, encode, imm8);
 }
 
+void Assembler::eshldl(Register dst, Register src1, Register src2, int8_t imm8, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(src1->encoding(), dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int24(0x24, (0xC0 | encode), imm8);
+}
+
 void Assembler::shrdl(Register dst, Register src) {
   int encode = prefix_and_encode(src->encoding(), dst->encoding(), true /* is_map1 */);
   emit_opcode_prefix_and_encoding((unsigned char)0xAD, 0xC0, encode);
 }
 
+void Assembler::eshrdl(Register dst, Register src1, Register src2, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(src1->encoding(), dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16(0xAD, (0xC0 | encode));
+}
+
 void Assembler::shrdl(Register dst, Register src, int8_t imm8) {
   int encode = prefix_and_encode(src->encoding(), dst->encoding(), true /* is_map1 */);
   emit_opcode_prefix_and_encoding((unsigned char)0xAC, 0xC0, encode, imm8);
 }
 
+void Assembler::eshrdl(Register dst, Register src1, Register src2, int8_t imm8, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(src1->encoding(), dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int24(0x2C, (0xC0 | encode), imm8);
+}
+
 #ifdef _LP64
 void Assembler::shldq(Register dst, Register src, int8_t imm8) {
   int encode = prefixq_and_encode(src->encoding(), dst->encoding(), true /* is_map1 */);
   emit_opcode_prefix_and_encoding((unsigned char)0xA4, 0xC0, encode, imm8);
 }
 
+void Assembler::eshldq(Register dst, Register src1, Register src2, int8_t imm8, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(src1->encoding(), dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int24(0x24, (0xC0 | encode), imm8);
+}
+
 void Assembler::shrdq(Register dst, Register src, int8_t imm8) {
   int encode = prefixq_and_encode(src->encoding(), dst->encoding(), true /* is_map1 */);
   emit_opcode_prefix_and_encoding((unsigned char)0xAC, 0xC0, encode, imm8);
 }
+
+void Assembler::eshrdq(Register dst, Register src1, Register src2, int8_t imm8, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(src1->encoding(), dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int24(0x2C, (0xC0 | encode), imm8);
+}
 #endif
 
 // copies a single word from [esi] to [edi]
@@ -6472,6 +7032,14 @@ void Assembler::subl(Address dst, int32_t imm32) {
   emit_arith_operand(0x81, rbp, dst, imm32);
 }
 
+void Assembler::esubl(Register dst, Address src, int32_t imm32, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_arith_operand(0x81, rbp, src, imm32);
+}
+
 void Assembler::subl(Address dst, Register src) {
   InstructionMark im(this);
   prefix(dst, src);
@@ -6479,17 +7047,38 @@ void Assembler::subl(Address dst, Register src) {
   emit_operand(src, dst, 0);
 }
 
+void Assembler::esubl(Register dst, Address src1, Register src2, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src1, dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0x29);
+  emit_operand(src2, src1, 0);
+}
+
 void Assembler::subl(Register dst, int32_t imm32) {
   prefix(dst);
   emit_arith(0x81, 0xE8, dst, imm32);
 }
 
+void Assembler::esubl(Register dst, Register src, int32_t imm32, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  (void) evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_arith(0x81, 0xE8, src, imm32);
+}
+
 // Force generation of a 4 byte immediate value even if it fits into 8bit
 void Assembler::subl_imm32(Register dst, int32_t imm32) {
   prefix(dst);
   emit_arith_imm32(0x81, 0xE8, dst, imm32);
 }
 
+void Assembler::esubl_imm32(Register dst, Register src, int32_t imm32, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  (void) evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_arith_imm32(0x81, 0xE8, src, imm32);
+}
+
 void Assembler::subl(Register dst, Address src) {
   InstructionMark im(this);
   prefix(src, dst);
@@ -6497,11 +7086,27 @@ void Assembler::subl(Register dst, Address src) {
   emit_operand(dst, src, 0);
 }
 
+void Assembler::esubl(Register dst, Register src1, Address src2, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src2, dst->encoding(), src1->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0x2B);
+  emit_operand(src1, src2, 0);
+}
+
 void Assembler::subl(Register dst, Register src) {
   (void) prefix_and_encode(dst->encoding(), src->encoding());
   emit_arith(0x2B, 0xC0, dst, src);
 }
 
+void Assembler::esubl(Register dst, Register src2, Register src1, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  (void) evex_prefix_and_encode_ndd(src1->encoding(), dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  // opcode matches gcc
+  emit_arith(0x29, 0xC0, src1, src2);
+}
+
 void Assembler::subsd(XMMRegister dst, XMMRegister src) {
   NOT_LP64(assert(VM_Version::supports_sse2(), ""));
   InstructionAttr attributes(AVX_128bit, /* rex_w */ VM_Version::supports_evex(), /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
@@ -6605,6 +7210,13 @@ void Assembler::tzcntl(Register dst, Register src) {
   emit_opcode_prefix_and_encoding((unsigned char)0xBC, 0xC0, encode);
 }
 
+void Assembler::etzcntl(Register dst, Register src, bool no_flags) {
+  assert(VM_Version::supports_bmi1(), "tzcnt instruction not supported");
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_nf(dst->encoding(), 0, src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xF4, (0xC0 | encode));
+}
+
 void Assembler::tzcntl(Register dst, Address src) {
   assert(VM_Version::supports_bmi1(), "tzcnt instruction not supported");
   InstructionMark im(this);
@@ -6614,6 +7226,16 @@ void Assembler::tzcntl(Register dst, Address src) {
   emit_operand(dst, src, 0);
 }
 
+void Assembler::etzcntl(Register dst, Address src, bool no_flags) {
+  assert(VM_Version::supports_bmi1(), "tzcnt instruction not supported");
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_nf(src, 0, dst->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8((unsigned char)0xF4);
+  emit_operand(dst, src, 0);
+}
+
 void Assembler::tzcntq(Register dst, Register src) {
   assert(VM_Version::supports_bmi1(), "tzcnt instruction not supported");
   emit_int8((unsigned char)0xF3);
@@ -6621,6 +7243,13 @@ void Assembler::tzcntq(Register dst, Register src) {
   emit_opcode_prefix_and_encoding((unsigned char)0xBC, 0xC0, encode);
 }
 
+void Assembler::etzcntq(Register dst, Register src, bool no_flags) {
+  assert(VM_Version::supports_bmi1(), "tzcnt instruction not supported");
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_nf(dst->encoding(), 0, src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xF4, (0xC0 | encode));
+}
+
 void Assembler::tzcntq(Register dst, Address src) {
   assert(VM_Version::supports_bmi1(), "tzcnt instruction not supported");
   InstructionMark im(this);
@@ -6630,6 +7259,16 @@ void Assembler::tzcntq(Register dst, Address src) {
   emit_operand(dst, src, 0);
 }
 
+void Assembler::etzcntq(Register dst, Address src, bool no_flags) {
+  assert(VM_Version::supports_bmi1(), "tzcnt instruction not supported");
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_nf(src, 0, dst->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8((unsigned char)0xF4);
+  emit_operand(dst, src, 0);
+}
+
 void Assembler::ucomisd(XMMRegister dst, Address src) {
   NOT_LP64(assert(VM_Version::supports_sse2(), ""));
   InstructionMark im(this);
@@ -6749,11 +7388,25 @@ void Assembler::xorl(Address dst, int32_t imm32) {
   emit_arith_operand(0x81, as_Register(6), dst, imm32);
 }
 
+void Assembler::exorl(Register dst, Address src, int32_t imm32, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_arith_operand(0x81, as_Register(6), src, imm32);
+}
+
 void Assembler::xorl(Register dst, int32_t imm32) {
   prefix(dst);
   emit_arith(0x81, 0xF0, dst, imm32);
 }
 
+void Assembler::exorl(Register dst, Register src, int32_t imm32, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_arith(0x81, 0xF0, src, imm32);
+}
+
 void Assembler::xorl(Register dst, Address src) {
   InstructionMark im(this);
   prefix(src, dst);
@@ -6761,11 +7414,27 @@ void Assembler::xorl(Register dst, Address src) {
   emit_operand(dst, src, 0);
 }
 
+void Assembler::exorl(Register dst, Register src1, Address src2, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src2, dst->encoding(), src1->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0x33);
+  emit_operand(src1, src2, 0);
+}
+
 void Assembler::xorl(Register dst, Register src) {
   (void) prefix_and_encode(dst->encoding(), src->encoding());
   emit_arith(0x33, 0xC0, dst, src);
 }
 
+void Assembler::exorl(Register dst, Register src1, Register src2, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  (void) evex_prefix_and_encode_ndd(src1->encoding(), dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  // opcode matches gcc
+  emit_arith(0x31, 0xC0, src1, src2);
+}
+
 void Assembler::xorl(Address dst, Register src) {
   InstructionMark im(this);
   prefix(dst, src);
@@ -6773,6 +7442,15 @@ void Assembler::xorl(Address dst, Register src) {
   emit_operand(src, dst, 0);
 }
 
+void Assembler::exorl(Register dst, Address src1, Register src2, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src1, dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0x31);
+  emit_operand(src2, src1, 0);
+}
+
 void Assembler::xorb(Register dst, Address src) {
   InstructionMark im(this);
   prefix(src, dst);
@@ -6780,6 +7458,15 @@ void Assembler::xorb(Register dst, Address src) {
   emit_operand(dst, src, 0);
 }
 
+void Assembler::exorb(Register dst, Register src1, Address src2, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_8bit);
+  evex_prefix_ndd(src2, dst->encoding(), src1->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0x32);
+  emit_operand(src1, src2, 0);
+}
+
 void Assembler::xorb(Address dst, Register src) {
   InstructionMark im(this);
   prefix(dst, src, true);
@@ -6787,6 +7474,15 @@ void Assembler::xorb(Address dst, Register src) {
   emit_operand(src, dst, 0);
 }
 
+void Assembler::exorb(Register dst, Address src1, Register src2, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_8bit);
+  evex_prefix_ndd(src1, dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0x30);
+  emit_operand(src2, src1, 0);
+}
+
 void Assembler::xorw(Register dst, Address src) {
   InstructionMark im(this);
   emit_int8(0x66);
@@ -6795,6 +7491,16 @@ void Assembler::xorw(Register dst, Address src) {
   emit_operand(dst, src, 0);
 }
 
+void Assembler::exorw(Register dst, Register src1, Address src2, bool no_flags) {
+  InstructionMark im(this);
+  emit_int8(0x66);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src2, dst->encoding(), src1->encoding(), VEX_SIMD_66, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0x33);
+  emit_operand(src1, src2, 0);
+}
+
 // AVX 3-operands scalar float-point arithmetic instructions
 
 void Assembler::vaddsd(XMMRegister dst, XMMRegister nds, Address src) {
@@ -11378,6 +12084,12 @@ void Assembler::decl(Register dst) {
  emit_int8(0x48 | dst->encoding());
 }
 
+void Assembler::edecl(Register dst, Register src, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  (void) evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0x48 | src->encoding());
+}
+
 // 64bit doesn't use the x87
 
 void Assembler::fabs() {
@@ -11816,7 +12528,7 @@ void Assembler::vex_prefix(bool vex_r, bool vex_b, bool vex_x, int nds_enc, VexS
 
 // This is a 4 byte encoding
 void Assembler::evex_prefix(bool vex_r, bool vex_b, bool vex_x, bool evex_r, bool eevex_b, bool evex_v,
-                       bool eevex_x, int nds_enc, VexSimdPrefix pre, VexOpcode opc) {
+                       bool eevex_x, int nds_enc, VexSimdPrefix pre, VexOpcode opc, bool no_flags) {
   // EVEX 0x62 prefix
   // byte1 = EVEX_4bytes;
 
@@ -11842,11 +12554,17 @@ void Assembler::evex_prefix(bool vex_r, bool vex_b, bool vex_x, bool evex_r, boo
   // of form {66, F3, F2}
   byte3 |= pre;
 
-  // P2: byte 4 as zL'Lbv'aaa
-  // kregs are implemented in the low 3 bits as aaa
-  int byte4 = (_attributes->is_no_reg_mask()) ?
-              0 :
-              _attributes->get_embedded_opmask_register_specifier();
+  // P2: byte 4 as zL'Lbv'aaa or 00LXVF00 where V = V4, X(extended context) = ND and F = NF (no flags)
+  int byte4 = 0;
+  if (no_flags) {
+    assert(_attributes->is_no_reg_mask(), "mask register not supported with no_flags");
+    byte4 |= 0x4;
+  } else {
+    // kregs are implemented in the low 3 bits as aaa
+    byte4 = (_attributes->is_no_reg_mask()) ?
+                0 :
+                _attributes->get_embedded_opmask_register_specifier();
+  }
   // EVEX.v` for extending EVEX.vvvv or VIDX
   byte4 |= (evex_v ? 0: EVEX_V);
   // third EXEC.b for broadcast actions
@@ -11861,11 +12579,12 @@ void Assembler::evex_prefix(bool vex_r, bool vex_b, bool vex_x, bool evex_r, boo
   emit_int32(EVEX_4bytes, byte2, byte3, byte4);
 }
 
-void Assembler::vex_prefix(Address adr, int nds_enc, int xreg_enc, VexSimdPrefix pre, VexOpcode opc, InstructionAttr *attributes) {
-  if (adr.base_needs_rex2() || adr.index_needs_rex2()) {
+void Assembler::vex_prefix(Address adr, int nds_enc, int xreg_enc, VexSimdPrefix pre, VexOpcode opc, InstructionAttr *attributes, bool nds_is_ndd, bool no_flags) {
+  if (adr.base_needs_rex2() || adr.index_needs_rex2() || nds_is_ndd || no_flags) {
     assert(UseAPX, "APX features not enabled");
   }
-  bool is_extended = adr.base_needs_rex2() || adr.index_needs_rex2() || nds_enc >= 16 || xreg_enc >= 16;
+  if (nds_is_ndd) attributes->set_extended_context();
+  bool is_extended = adr.base_needs_rex2() || adr.index_needs_rex2() || nds_enc >= 16 || xreg_enc >= 16 || nds_is_ndd;
   bool vex_r = (xreg_enc & 8) == 8;
   bool vex_b = adr.base_needs_rex();
   bool vex_x;
@@ -11908,7 +12627,7 @@ void Assembler::vex_prefix(Address adr, int nds_enc, int xreg_enc, VexSimdPrefix
     bool eevex_x = adr.index_needs_rex2();
     bool eevex_b = adr.base_needs_rex2();
     attributes->set_is_evex_instruction();
-    evex_prefix(vex_r, vex_b, vex_x, evex_r, eevex_b, evex_v, eevex_x, nds_enc, pre, opc);
+    evex_prefix(vex_r, vex_b, vex_x, evex_r, eevex_b, evex_v, eevex_x, nds_enc, pre, opc, no_flags);
   } else {
     if (UseAVX > 2 && attributes->is_rex_vex_w_reverted()) {
       attributes->set_rex_vex_w(false);
@@ -11917,10 +12636,21 @@ void Assembler::vex_prefix(Address adr, int nds_enc, int xreg_enc, VexSimdPrefix
   }
 }
 
-int Assembler::vex_prefix_and_encode(int dst_enc, int nds_enc, int src_enc, VexSimdPrefix pre, VexOpcode opc, InstructionAttr *attributes, bool src_is_gpr) {
-  if (src_is_gpr && src_enc >= 16) {
+void Assembler::evex_prefix_ndd(Address adr, int ndd_enc, int xreg_enc, VexSimdPrefix pre, VexOpcode opc, InstructionAttr *attributes, bool no_flags) {
+  attributes->set_is_evex_instruction();
+  vex_prefix(adr, ndd_enc, xreg_enc, pre, opc, attributes, /* nds_is_ndd */ true, no_flags);
+}
+
+void Assembler::evex_prefix_nf(Address adr, int ndd_enc, int xreg_enc, VexSimdPrefix pre, VexOpcode opc, InstructionAttr *attributes, bool no_flags) {
+  attributes->set_is_evex_instruction();
+  vex_prefix(adr, ndd_enc, xreg_enc, pre, opc, attributes, /* nds_is_ndd */ false, no_flags);
+}
+
+int Assembler::vex_prefix_and_encode(int dst_enc, int nds_enc, int src_enc, VexSimdPrefix pre, VexOpcode opc, InstructionAttr *attributes, bool src_is_gpr, bool nds_is_ndd, bool no_flags) {
+  if (nds_is_ndd || no_flags || (src_is_gpr && src_enc >= 16)) {
     assert(UseAPX, "APX features not enabled");
   }
+  if (nds_is_ndd) attributes->set_extended_context();
   bool is_extended = dst_enc >= 16 || nds_enc >= 16 || src_enc >=16;
   bool vex_r = (dst_enc & 8) == 8;
   bool vex_b = (src_enc & 8) == 8;
@@ -11962,7 +12692,7 @@ int Assembler::vex_prefix_and_encode(int dst_enc, int nds_enc, int src_enc, VexS
     // can use vex_x as bank extender on rm encoding
     vex_x = (src_enc >= 16) && !src_is_gpr;
     attributes->set_is_evex_instruction();
-    evex_prefix(vex_r, vex_b, vex_x, evex_r, evex_b, evex_v, false /*eevex_x*/, nds_enc, pre, opc);
+    evex_prefix(vex_r, vex_b, vex_x, evex_r, evex_b, evex_v, false /*eevex_x*/, nds_enc, pre, opc, no_flags);
   } else {
     if (UseAVX > 2 && attributes->is_rex_vex_w_reverted()) {
       attributes->set_rex_vex_w(false);
@@ -11974,6 +12704,18 @@ int Assembler::vex_prefix_and_encode(int dst_enc, int nds_enc, int src_enc, VexS
   return (((dst_enc & 7) << 3) | (src_enc & 7));
 }
 
+int Assembler::evex_prefix_and_encode_ndd(int dst_enc, int nds_enc, int src_enc, VexSimdPrefix pre, VexOpcode opc,
+                           InstructionAttr *attributes, bool no_flags) {
+  attributes->set_is_evex_instruction();
+  return vex_prefix_and_encode(dst_enc, nds_enc, src_enc, pre, opc, attributes, /* src_is_gpr */ true, /* nds_is_ndd */ true, no_flags);
+}
+
+int Assembler::evex_prefix_and_encode_nf(int dst_enc, int nds_enc, int src_enc, VexSimdPrefix pre, VexOpcode opc,
+                           InstructionAttr *attributes, bool no_flags) {
+  attributes->set_is_evex_instruction();
+  return vex_prefix_and_encode(dst_enc, nds_enc, src_enc, pre, opc, attributes, /* src_is_gpr */ true, /* nds_is_ndd */ false, no_flags);
+}
+
 void Assembler::simd_prefix(XMMRegister xreg, XMMRegister nds, Address adr, VexSimdPrefix pre,
                             VexOpcode opc, InstructionAttr *attributes) {
   if (UseAVX > 0) {
@@ -12818,6 +13560,12 @@ void Assembler::incl(Register dst) {
   emit_int8(0x40 | dst->encoding());
 }
 
+void Assembler::eincl(Register dst, Register src, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  (void) evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0x40 | src->encoding());
+}
+
 void Assembler::lea(Register dst, Address src) {
   leal(dst, src);
 }
@@ -13442,28 +14190,67 @@ void Assembler::addq(Address dst, int32_t imm32) {
   emit_arith_operand(0x81, rax, dst, imm32);
 }
 
+void Assembler::eaddq(Register dst, Address src, int32_t imm32, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_arith_operand(0x81, rax, src, imm32);
+}
+
 void Assembler::addq(Address dst, Register src) {
   InstructionMark im(this);
   emit_prefix_and_int8(get_prefixq(dst, src), 0x01);
   emit_operand(src, dst, 0);
 }
 
+void Assembler::eaddq(Register dst, Address src1, Register src2, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src1, dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0x01);
+  emit_operand(src2, src1, 0);
+}
+
 void Assembler::addq(Register dst, int32_t imm32) {
   (void) prefixq_and_encode(dst->encoding());
   emit_arith(0x81, 0xC0, dst, imm32);
 }
 
+void Assembler::eaddq(Register dst, Register src, int32_t imm32, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  (void) evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_arith(0x81, 0xC0, src, imm32);
+}
+
 void Assembler::addq(Register dst, Address src) {
   InstructionMark im(this);
   emit_prefix_and_int8(get_prefixq(src, dst), 0x03);
   emit_operand(dst, src, 0);
 }
 
+void Assembler::eaddq(Register dst, Register src1, Address src2, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src2, dst->encoding(), src1->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0x03);
+  emit_operand(src1, src2, 0);
+}
+
 void Assembler::addq(Register dst, Register src) {
   (void) prefixq_and_encode(dst->encoding(), src->encoding());
   emit_arith(0x03, 0xC0, dst, src);
 }
 
+void Assembler::eaddq(Register dst, Register src1, Register src2, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  (void) evex_prefix_and_encode_ndd(src1->encoding(), dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  // opcode matches gcc
+  emit_arith(0x01, 0xC0, src1, src2);
+}
+
 void Assembler::adcxq(Register dst, Register src) {
   //assert(VM_Version::supports_adx(), "adx instructions not supported");
   if (needs_rex2(dst, src)) {
@@ -13480,6 +14267,12 @@ void Assembler::adcxq(Register dst, Register src) {
   }
 }
 
+void Assembler::eadcxq(Register dst, Register src1, Register src2) {
+    InstructionAttr attributes(AVX_128bit, /* rex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+    int encode = evex_prefix_and_encode_ndd(src1->encoding(), dst->encoding(), src2->encoding(), VEX_SIMD_66, VEX_OPCODE_0F_3C, &attributes);
+    emit_int16((unsigned char)0x66, (0xC0 | encode));
+}
+
 void Assembler::adoxq(Register dst, Register src) {
   //assert(VM_Version::supports_adx(), "adx instructions not supported");
   if (needs_rex2(dst, src)) {
@@ -13495,34 +14288,80 @@ void Assembler::adoxq(Register dst, Register src) {
                (0xC0 | encode));
   }
 }
-void Assembler::andq(Address dst, int32_t imm32) {
+
+void Assembler::eadoxq(Register dst, Register src1, Register src2) {
+    InstructionAttr attributes(AVX_128bit, /* rex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+    int encode = evex_prefix_and_encode_ndd(src1->encoding(), dst->encoding(), src2->encoding(), VEX_SIMD_F3, VEX_OPCODE_0F_3C, &attributes);
+    emit_int16((unsigned char)0x66, (0xC0 | encode));
+}
+
+void Assembler::andq(Address dst, int32_t imm32) {
   InstructionMark im(this);
   prefixq(dst);
   emit_arith_operand(0x81, as_Register(4), dst, imm32);
 }
 
+void Assembler::eandq(Register dst, Address src, int32_t imm32, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_arith_operand(0x81, as_Register(4), src, imm32);
+}
+
 void Assembler::andq(Register dst, int32_t imm32) {
   (void) prefixq_and_encode(dst->encoding());
   emit_arith(0x81, 0xE0, dst, imm32);
 }
 
+void Assembler::eandq(Register dst, Register src, int32_t imm32, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_arith(0x81, 0xE0, src, imm32);
+}
+
 void Assembler::andq(Register dst, Address src) {
   InstructionMark im(this);
   emit_prefix_and_int8(get_prefixq(src, dst), 0x23);
   emit_operand(dst, src, 0);
 }
 
+void Assembler::eandq(Register dst, Register src1, Address src2, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src2, dst->encoding(), src1->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0x23);
+  emit_operand(src1, src2, 0);
+}
+
 void Assembler::andq(Register dst, Register src) {
   (void) prefixq_and_encode(dst->encoding(), src->encoding());
   emit_arith(0x23, 0xC0, dst, src);
 }
 
+void Assembler::eandq(Register dst, Register src1, Register src2, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  (void) evex_prefix_and_encode_ndd(src1->encoding(), dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  // opcode matches gcc
+  emit_arith(0x21, 0xC0, src1, src2);
+}
+
 void Assembler::andq(Address dst, Register src) {
   InstructionMark im(this);
   emit_prefix_and_int8(get_prefixq(dst, src), 0x21);
   emit_operand(src, dst, 0);
 }
 
+void Assembler::eandq(Register dst, Address src1, Register src2, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src1, dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0x21);
+  emit_operand(src2, src1, 0);
+}
+
 void Assembler::andnq(Register dst, Register src1, Register src2) {
   assert(VM_Version::supports_bmi1(), "bit manipulation instructions not supported");
   InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
@@ -13656,6 +14495,12 @@ void Assembler::cmovq(Condition cc, Register dst, Register src) {
   emit_opcode_prefix_and_encoding((0x40 | cc), 0xC0, encode);
 }
 
+void Assembler::ecmovq(Condition cc, Register dst, Register src1, Register src2) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(src1->encoding(), dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes);
+  emit_int16((0x40 | cc), (0xC0 | encode));
+}
+
 void Assembler::cmovq(Condition cc, Register dst, Address src) {
   InstructionMark im(this);
   int prefix = get_prefixq(src, dst, true /* is_map1 */);
@@ -13663,6 +14508,15 @@ void Assembler::cmovq(Condition cc, Register dst, Address src) {
   emit_operand(dst, src, 0);
 }
 
+void Assembler::ecmovq(Condition cc, Register dst, Register src1, Address src2) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src2, dst->encoding(), src1->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes);
+  emit_int8((0x40 | cc));
+  emit_operand(src1, src2, 0);
+}
+
 void Assembler::cmpq(Address dst, int32_t imm32) {
   InstructionMark im(this);
   prefixq(dst);
@@ -13764,6 +14618,12 @@ void Assembler::decl(Register dst) {
   emit_int16((unsigned char)0xFF, (0xC8 | encode));
 }
 
+void Assembler::edecl(Register dst, Register src, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xFF, (0xC8 | encode));
+}
+
 void Assembler::decq(Register dst) {
   // Don't use it directly. Use MacroAssembler::decrementq() instead.
   // Use two-byte form (one-byte from is a REX prefix in 64-bit mode)
@@ -13771,6 +14631,12 @@ void Assembler::decq(Register dst) {
   emit_int16((unsigned char)0xFF, 0xC8 | encode);
 }
 
+void Assembler::edecq(Register dst, Register src, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xFF, (0xC8 | encode));
+}
+
 void Assembler::decq(Address dst) {
   // Don't use it directly. Use MacroAssembler::decrementq() instead.
   InstructionMark im(this);
@@ -13778,6 +14644,15 @@ void Assembler::decq(Address dst) {
   emit_operand(rcx, dst, 0);
 }
 
+void Assembler::edecq(Register dst, Address src, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8((unsigned char)0xFF);
+  emit_operand(rcx, src, 0);
+}
+
 // can't use REX2
 void Assembler::fxrstor(Address src) {
   InstructionMark im(this);
@@ -13811,21 +14686,51 @@ void Assembler::idivq(Register src) {
   emit_int16((unsigned char)0xF7, (0xF8 | encode));
 }
 
+void Assembler::eidivq(Register src, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_nf(0, 0, src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xF7, (0xF8 | encode));
+}
+
 void Assembler::divq(Register src) {
   int encode = prefixq_and_encode(src->encoding());
   emit_int16((unsigned char)0xF7, (0xF0 | encode));
 }
 
+void Assembler::edivq(Register src, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_nf(0, 0, src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xF7, (0xF0 | encode));
+}
+
 void Assembler::imulq(Register dst, Register src) {
   int encode = prefixq_and_encode(dst->encoding(), src->encoding(), true /* is_map1 */);
   emit_opcode_prefix_and_encoding((unsigned char)0xAF, 0xC0, encode);
 }
 
+void Assembler::eimulq(Register dst, Register src, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_nf(dst->encoding(), 0, src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xAF, (0xC0 | encode));
+}
+
+void Assembler::eimulq(Register dst, Register src1, Register src2, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(src1->encoding(), dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xAF, (0xC0 | encode));
+}
+
 void Assembler::imulq(Register src) {
   int encode = prefixq_and_encode(src->encoding());
   emit_int16((unsigned char)0xF7, (0xE8 | encode));
 }
 
+void Assembler::eimulq(Register src, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_nf(0, 0, src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xF7, (0xE8 | encode));
+}
+
 void Assembler::imulq(Register dst, Address src, int32_t value) {
   InstructionMark im(this);
   prefixq(src, dst);
@@ -13840,6 +14745,22 @@ void Assembler::imulq(Register dst, Address src, int32_t value) {
   }
 }
 
+void Assembler::eimulq(Register dst, Address src, int32_t value, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_nf(src, 0, dst->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  if (is8bit(value)) {
+    emit_int8((unsigned char)0x6B);
+    emit_operand(dst, src, 1);
+    emit_int8(value);
+  } else {
+    emit_int8((unsigned char)0x69);
+    emit_operand(dst, src, 4);
+    emit_int32(value);
+  }
+}
+
 void Assembler::imulq(Register dst, Register src, int value) {
   int encode = prefixq_and_encode(dst->encoding(), src->encoding());
   if (is8bit(value)) {
@@ -13850,6 +14771,17 @@ void Assembler::imulq(Register dst, Register src, int value) {
   }
 }
 
+void Assembler::eimulq(Register dst, Register src, int value, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = vex_prefix_and_encode(dst->encoding(), 0, src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, /* src_is_gpr */ true, /* nds_is_ndd */ false, no_flags);
+  if (is8bit(value)) {
+    emit_int24(0x6B, (0xC0 | encode), (value & 0xFF));
+  } else {
+    emit_int16(0x69, (0xC0 | encode));
+    emit_int32(value);
+  }
+}
+
 void Assembler::imulq(Register dst, Address src) {
   InstructionMark im(this);
   int prefix = get_prefixq(src, dst, true /* is_map1 */);
@@ -13857,6 +14789,23 @@ void Assembler::imulq(Register dst, Address src) {
   emit_operand(dst, src, 0);
 }
 
+void Assembler::eimulq(Register dst, Address src, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  vex_prefix(src, 0, dst->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F, &attributes, /* nds_is_ndd */ false, no_flags);
+  emit_int8((unsigned char)0xAF);
+  emit_operand(dst, src, 0);
+}
+
+void Assembler::eimulq(Register dst, Register src1, Address src2, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_32bit);
+  evex_prefix_ndd(src2, dst->encoding(), src1->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8((unsigned char)0xAF);
+  emit_operand(src1, src2, 0);
+}
+
 void Assembler::incl(Register dst) {
   // Don't use it directly. Use MacroAssembler::incrementl() instead.
   // Use two-byte form (one-byte from is a REX prefix in 64-bit mode)
@@ -13864,6 +14813,15 @@ void Assembler::incl(Register dst) {
   emit_int16((unsigned char)0xFF, (0xC0 | encode));
 }
 
+void Assembler::eincl(Register dst, Register src, bool no_flags) {
+  // Don't use it directly. Use MacroAssembler::incrementl() instead.
+  // Use two-byte form (one-byte from is a REX prefix in 64-bit mode)
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  // int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xFF, (0xC0 | encode));
+}
+
 void Assembler::incq(Register dst) {
   // Don't use it directly. Use MacroAssembler::incrementq() instead.
   // Use two-byte form (one-byte from is a REX prefix in 64-bit mode)
@@ -13871,6 +14829,14 @@ void Assembler::incq(Register dst) {
   emit_int16((unsigned char)0xFF, (0xC0 | encode));
 }
 
+void Assembler::eincq(Register dst, Register src, bool no_flags) {
+  // Don't use it directly. Use MacroAssembler::incrementq() instead.
+  // Use two-byte form (one-byte from is a REX prefix in 64-bit mode)
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xFF, (0xC0 | encode));
+}
+
 void Assembler::incq(Address dst) {
   // Don't use it directly. Use MacroAssembler::incrementq() instead.
   InstructionMark im(this);
@@ -13878,6 +14844,16 @@ void Assembler::incq(Address dst) {
   emit_operand(rax, dst, 0);
 }
 
+void Assembler::eincq(Register dst, Address src, bool no_flags) {
+  // Don't use it directly. Use MacroAssembler::incrementq() instead.
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8((unsigned char) 0xFF);
+  emit_operand(rax, src, 0);
+}
+
 void Assembler::lea(Register dst, Address src) {
   leaq(dst, src);
 }
@@ -13946,6 +14922,13 @@ void Assembler::lzcntq(Register dst, Register src) {
   emit_opcode_prefix_and_encoding((unsigned char)0xBD, 0xC0, encode);
 }
 
+void Assembler::elzcntq(Register dst, Register src, bool no_flags) {
+  assert(VM_Version::supports_lzcnt(), "encoding is treated as BSR");
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_nf(dst->encoding(), 0, src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xF5, (0xC0 | encode));
+}
+
 void Assembler::lzcntq(Register dst, Address src) {
   assert(VM_Version::supports_lzcnt(), "encoding is treated as BSR");
   InstructionMark im(this);
@@ -13955,6 +14938,16 @@ void Assembler::lzcntq(Register dst, Address src) {
   emit_operand(dst, src, 0);
 }
 
+void Assembler::elzcntq(Register dst, Address src, bool no_flags) {
+  assert(VM_Version::supports_lzcnt(), "encoding is treated as BSR");
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_nf(src, 0, dst->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8((unsigned char)0xF5);
+  emit_operand(dst, src, 0);
+}
+
 void Assembler::movdq(XMMRegister dst, Register src) {
   // table D-1 says MMX/SSE2
   NOT_LP64(assert(VM_Version::supports_sse2(), ""));
@@ -14077,11 +15070,26 @@ void Assembler::mulq(Address src) {
   emit_operand(rsp, src, 0);
 }
 
+void Assembler::emulq(Address src, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_nf(src, 0, 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0xF7);
+  emit_operand(rsp, src, 0);
+}
+
 void Assembler::mulq(Register src) {
   int encode = prefixq_and_encode(src->encoding());
   emit_int16((unsigned char)0xF7, (0xE0 | encode));
 }
 
+void Assembler::emulq(Register src, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_nf(0, 0, src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xF7, (0xE0 | encode));
+}
+
 void Assembler::mulxq(Register dst1, Register dst2, Register src) {
   assert(VM_Version::supports_bmi2(), "bit manipulation instructions not supported");
   InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
@@ -14094,17 +15102,38 @@ void Assembler::negq(Register dst) {
   emit_int16((unsigned char)0xF7, (0xD8 | encode));
 }
 
+void Assembler::enegq(Register dst, Register src, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xF7, (0xD8 | encode));
+}
+
 void Assembler::negq(Address dst) {
   InstructionMark im(this);
   emit_prefix_and_int8(get_prefixq(dst), (unsigned char)0xF7);
   emit_operand(as_Register(3), dst, 0);
 }
 
+void Assembler::enegq(Register dst, Address src, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8((unsigned char)0xF7);
+  emit_operand(as_Register(3), src, 0);
+}
+
 void Assembler::notq(Register dst) {
   int encode = prefixq_and_encode(dst->encoding());
   emit_int16((unsigned char)0xF7, (0xD0 | encode));
 }
 
+void Assembler::enotq(Register dst, Register src) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes);
+  emit_int16((unsigned char)0xF7, (0xD0 | encode));
+}
+
 void Assembler::btq(Register dst, Register src) {
   int encode = prefixq_and_encode(src->encoding(), dst->encoding(), true /* is_map1 */);
   emit_opcode_prefix_and_encoding((unsigned char)0xA3, 0xC0, encode);
@@ -14141,33 +15170,78 @@ void Assembler::orq(Address dst, int32_t imm32) {
   emit_arith_operand(0x81, as_Register(1), dst, imm32);
 }
 
+void Assembler::eorq(Register dst, Address src, int32_t imm32, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_arith_operand(0x81, as_Register(1), src, imm32);
+}
+
 void Assembler::orq(Address dst, Register src) {
   InstructionMark im(this);
   emit_prefix_and_int8(get_prefixq(dst, src), (unsigned char)0x09);
   emit_operand(src, dst, 0);
 }
 
+void Assembler::eorq(Register dst, Address src1, Register src2, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src1, dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0x09);
+  emit_operand(src2, src1, 0);
+}
+
 void Assembler::orq(Register dst, int32_t imm32) {
   (void) prefixq_and_encode(dst->encoding());
   emit_arith(0x81, 0xC8, dst, imm32);
 }
 
+void Assembler::eorq(Register dst, Register src, int32_t imm32, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_arith(0x81, 0xC8, src, imm32);
+}
+
 void Assembler::orq_imm32(Register dst, int32_t imm32) {
   (void) prefixq_and_encode(dst->encoding());
   emit_arith_imm32(0x81, 0xC8, dst, imm32);
 }
 
+void Assembler::eorq_imm32(Register dst, Register src, int32_t imm32, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  (void) evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_arith_imm32(0x81, 0xC8, src, imm32);
+}
+
 void Assembler::orq(Register dst, Address src) {
   InstructionMark im(this);
   emit_prefix_and_int8(get_prefixq(src, dst), 0x0B);
   emit_operand(dst, src, 0);
 }
 
+void Assembler::eorq(Register dst, Register src1, Address src2, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src2, dst->encoding(), src1->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0x0B);
+  emit_operand(src1, src2, 0);
+}
+
 void Assembler::orq(Register dst, Register src) {
   (void) prefixq_and_encode(dst->encoding(), src->encoding());
   emit_arith(0x0B, 0xC0, dst, src);
 }
 
+void Assembler::eorq(Register dst, Register src1, Register src2, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  (void) evex_prefix_and_encode_ndd(src1->encoding(), dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  // opcode matches gcc
+  emit_arith(0x09, 0xC0, src1, src2);
+}
+
 void Assembler::popcntq(Register dst, Address src) {
   assert(VM_Version::supports_popcnt(), "must support");
   InstructionMark im(this);
@@ -14176,6 +15250,16 @@ void Assembler::popcntq(Register dst, Address src) {
   emit_operand(dst, src, 0);
 }
 
+void Assembler::epopcntq(Register dst, Address src, bool no_flags) {
+  assert(VM_Version::supports_popcnt(), "must support");
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_nf(src, 0, dst->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8((unsigned char) 0x88);
+  emit_operand(dst, src, 0);
+}
+
 void Assembler::popcntq(Register dst, Register src) {
   assert(VM_Version::supports_popcnt(), "must support");
   emit_int8((unsigned char)0xF3);
@@ -14183,6 +15267,13 @@ void Assembler::popcntq(Register dst, Register src) {
   emit_opcode_prefix_and_encoding((unsigned char)0xB8, 0xC0, encode);
 }
 
+void Assembler::epopcntq(Register dst, Register src, bool no_flags) {
+  assert(VM_Version::supports_popcnt(), "must support");
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_nf(dst->encoding(), 0, src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0x88, (0xC0 | encode));
+}
+
 void Assembler::popq(Address dst) {
   InstructionMark im(this);
   emit_prefix_and_int8(get_prefixq(dst), (unsigned char)0x8F);
@@ -14401,6 +15492,17 @@ void Assembler::rclq(Register dst, int imm8) {
   }
 }
 
+void Assembler::erclq(Register dst, Register src, int imm8) {
+  assert(isShiftCount(imm8 >> 1), "illegal shift count");
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode =  evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes);
+  if (imm8 == 1) {
+    emit_int16((unsigned char)0xD1, (0xD0 | encode));
+  } else {
+    emit_int24((unsigned char)0xC1, (0xD0 | encode), imm8);
+  }
+}
+
 void Assembler::rcrq(Register dst, int imm8) {
   assert(isShiftCount(imm8 >> 1), "illegal shift count");
   int encode = prefixq_and_encode(dst->encoding());
@@ -14411,6 +15513,17 @@ void Assembler::rcrq(Register dst, int imm8) {
   }
 }
 
+void Assembler::ercrq(Register dst, Register src, int imm8) {
+  assert(isShiftCount(imm8 >> 1), "illegal shift count");
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode =  evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes);
+  if (imm8 == 1) {
+    emit_int16((unsigned char)0xD1, (0xD8 | encode));
+  } else {
+    emit_int24((unsigned char)0xC1, (0xD8 | encode), imm8);
+  }
+}
+
 void Assembler::rorxl(Register dst, Register src, int imm8) {
   assert(VM_Version::supports_bmi2(), "bit manipulation instructions not supported");
   InstructionAttr attributes(AVX_128bit, /* vex_w */ false, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
@@ -14462,12 +15575,38 @@ void Assembler::salq(Address dst, int imm8) {
   }
 }
 
+void Assembler::esalq(Register dst, Address src, int imm8, bool no_flags) {
+  InstructionMark im(this);
+  assert(isShiftCount(imm8 >> 1), "illegal shift count");
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  if (imm8 == 1) {
+    emit_int8((unsigned char)0xD1);
+    emit_operand(as_Register(4), src, 0);
+  }
+  else {
+    emit_int8((unsigned char)0xC1);
+    emit_operand(as_Register(4), src, 1);
+    emit_int8(imm8);
+  }
+}
+
 void Assembler::salq(Address dst) {
   InstructionMark im(this);
   emit_prefix_and_int8(get_prefixq(dst), (unsigned char)0xD3);
   emit_operand(as_Register(4), dst, 0);
 }
 
+void Assembler::esalq(Register dst, Address src, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8((unsigned char)0xD3);
+  emit_operand(as_Register(4), src, 0);
+}
+
 void Assembler::salq(Register dst, int imm8) {
   assert(isShiftCount(imm8 >> 1), "illegal shift count");
   int encode = prefixq_and_encode(dst->encoding());
@@ -14478,11 +15617,28 @@ void Assembler::salq(Register dst, int imm8) {
   }
 }
 
+void Assembler::esalq(Register dst, Register src, int imm8, bool no_flags) {
+  assert(isShiftCount(imm8 >> 1), "illegal shift count");
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  if (imm8 == 1) {
+    emit_int16((unsigned char)0xD1, (0xE0 | encode));
+  } else {
+    emit_int24((unsigned char)0xC1, (0xE0 | encode), imm8);
+  }
+}
+
 void Assembler::salq(Register dst) {
   int encode = prefixq_and_encode(dst->encoding());
   emit_int16((unsigned char)0xD3, (0xE0 | encode));
 }
 
+void Assembler::esalq(Register dst, Register src, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xD3, (0xE0 | encode));
+}
+
 void Assembler::sarq(Address dst, int imm8) {
   InstructionMark im(this);
   assert(isShiftCount(imm8 >> 1), "illegal shift count");
@@ -14497,12 +15653,38 @@ void Assembler::sarq(Address dst, int imm8) {
   }
 }
 
+void Assembler::esarq(Register dst, Address src, int imm8, bool no_flags) {
+  assert(isShiftCount(imm8 >> 1), "illegal shift count");
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  if (imm8 == 1) {
+    emit_int8((unsigned char)0xD1);
+    emit_operand(as_Register(7), src, 0);
+  }
+  else {
+    emit_int8((unsigned char)0xC1);
+    emit_operand(as_Register(7), src, 1);
+    emit_int8(imm8);
+  }
+}
+
 void Assembler::sarq(Address dst) {
   InstructionMark im(this);
   emit_prefix_and_int8(get_prefixq(dst), (unsigned char)0xD3);
   emit_operand(as_Register(7), dst, 0);
 }
 
+void Assembler::esarq(Register dst, Address src, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8((unsigned char)0xD3);
+  emit_operand(as_Register(7), src, 0);
+}
+
 void Assembler::sarq(Register dst, int imm8) {
   assert(isShiftCount(imm8 >> 1), "illegal shift count");
   int encode = prefixq_and_encode(dst->encoding());
@@ -14513,10 +15695,26 @@ void Assembler::sarq(Register dst, int imm8) {
   }
 }
 
+void Assembler::esarq(Register dst, Register src, int imm8, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  if (imm8 == 1) {
+    emit_int16((unsigned char)0xD1, (0xF8 | encode));
+  } else {
+    emit_int24((unsigned char)0xC1, (0xF8 | encode), imm8);
+  }
+}
+
 void Assembler::sarq(Register dst) {
   int encode = prefixq_and_encode(dst->encoding());
   emit_int16((unsigned char)0xD3, (0xF8 | encode));
 }
+
+void Assembler::esarq(Register dst, Register src, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xD3, (0xF8 | encode));
+}
 #endif
 
 void Assembler::sbbq(Address dst, int32_t imm32) {
@@ -14551,11 +15749,28 @@ void Assembler::shlq(Register dst, int imm8) {
   }
 }
 
+void Assembler::eshlq(Register dst, Register src, int imm8, bool no_flags) {
+  assert(isShiftCount(imm8 >> 1), "illegal shift count");
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  if (imm8 == 1 ) {
+    emit_int16((unsigned char)0xD1, (0xE0 | encode));
+  } else {
+    emit_int24((unsigned char)0xC1, (0xE0 | encode), imm8);
+  }
+}
+
 void Assembler::shlq(Register dst) {
   int encode = prefixq_and_encode(dst->encoding());
   emit_int16((unsigned char)0xD3, (0xE0 | encode));
 }
 
+void Assembler::eshlq(Register dst, Register src, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xD3, (0xE0 | encode));
+}
+
 void Assembler::shrq(Register dst, int imm8) {
   assert(isShiftCount(imm8 >> 1), "illegal shift count");
   int encode = prefixq_and_encode(dst->encoding());
@@ -14567,17 +15782,44 @@ void Assembler::shrq(Register dst, int imm8) {
   }
 }
 
+void Assembler::eshrq(Register dst, Register src, int imm8, bool no_flags) {
+  assert(isShiftCount(imm8 >> 1), "illegal shift count");
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  if (imm8 == 1) {
+    emit_int16((unsigned char)0xD1, (0xE8 | encode));
+  }
+  else {
+    emit_int24((unsigned char)0xC1, (0xE8 | encode), imm8);
+  }
+}
+
 void Assembler::shrq(Register dst) {
   int encode = prefixq_and_encode(dst->encoding());
   emit_int16((unsigned char)0xD3, 0xE8 | encode);
 }
 
+void Assembler::eshrq(Register dst, Register src, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  int encode = evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int16((unsigned char)0xD3, (0xE8 | encode));
+}
+
 void Assembler::shrq(Address dst) {
   InstructionMark im(this);
   emit_prefix_and_int8(get_prefixq(dst), (unsigned char)0xD3);
   emit_operand(as_Register(5), dst, 0);
 }
 
+void Assembler::eshrq(Register dst, Address src, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8((unsigned char)0xD3);
+  emit_operand(as_Register(5), src, 0);
+}
+
 void Assembler::shrq(Address dst, int imm8) {
   InstructionMark im(this);
   assert(isShiftCount(imm8 >> 1), "illegal shift count");
@@ -14592,40 +15834,102 @@ void Assembler::shrq(Address dst, int imm8) {
   }
 }
 
+void Assembler::eshrq(Register dst, Address src, int imm8, bool no_flags) {
+  InstructionMark im(this);
+  assert(isShiftCount(imm8 >> 1), "illegal shift count");
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  if (imm8 == 1) {
+    emit_int8((unsigned char)0xD1);
+    emit_operand(as_Register(5), src, 0);
+  }
+  else {
+    emit_int8((unsigned char)0xC1);
+    emit_operand(as_Register(5), src, 1);
+    emit_int8(imm8);
+  }
+}
+
 void Assembler::subq(Address dst, int32_t imm32) {
   InstructionMark im(this);
   prefixq(dst);
   emit_arith_operand(0x81, rbp, dst, imm32);
 }
 
+void Assembler::esubq(Register dst, Address src, int32_t imm32, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_arith_operand(0x81, rbp, src, imm32);
+}
+
 void Assembler::subq(Address dst, Register src) {
   InstructionMark im(this);
   emit_prefix_and_int8(get_prefixq(dst, src), 0x29);
   emit_operand(src, dst, 0);
 }
 
+void Assembler::esubq(Register dst, Address src1, Register src2, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src1, dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0x29);
+  emit_operand(src2, src1, 0);
+}
+
 void Assembler::subq(Register dst, int32_t imm32) {
   (void) prefixq_and_encode(dst->encoding());
   emit_arith(0x81, 0xE8, dst, imm32);
 }
 
+void Assembler::esubq(Register dst, Register src, int32_t imm32, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  (void) evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_arith(0x81, 0xE8, src, imm32);
+}
+
 // Force generation of a 4 byte immediate value even if it fits into 8bit
 void Assembler::subq_imm32(Register dst, int32_t imm32) {
   (void) prefixq_and_encode(dst->encoding());
   emit_arith_imm32(0x81, 0xE8, dst, imm32);
 }
 
+void Assembler::esubq_imm32(Register dst, Register src, int32_t imm32, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  (void) evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_arith_imm32(0x81, 0xE8, src, imm32);
+}
+
 void Assembler::subq(Register dst, Address src) {
   InstructionMark im(this);
   emit_prefix_and_int8(get_prefixq(src, dst), 0x2B);
   emit_operand(dst, src, 0);
 }
 
+void Assembler::esubq(Register dst, Register src1, Address src2, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src2, dst->encoding(), src1->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0x2B);
+  emit_operand(src1, src2, 0);
+}
+
 void Assembler::subq(Register dst, Register src) {
   (void) prefixq_and_encode(dst->encoding(), src->encoding());
   emit_arith(0x2B, 0xC0, dst, src);
 }
 
+void Assembler::esubq(Register dst, Register src1, Register src2, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  (void) evex_prefix_and_encode_ndd(src1->encoding(), dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  // opcode matches gcc
+  emit_arith(0x29, 0xC0, src1, src2);
+}
+
 void Assembler::testq(Address dst, int32_t imm32) {
   InstructionMark im(this);
   emit_prefix_and_int8(get_prefixq(dst), (unsigned char)0xF7);
@@ -14683,29 +15987,68 @@ void Assembler::xorq(Register dst, Register src) {
   emit_arith(0x33, 0xC0, dst, src);
 }
 
+void Assembler::exorq(Register dst, Register src1, Register src2, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  (void) evex_prefix_and_encode_ndd(src1->encoding(), dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  // opcode matches gcc
+  emit_arith(0x31, 0xC0, src1, src2);
+}
+
 void Assembler::xorq(Register dst, Address src) {
   InstructionMark im(this);
   emit_prefix_and_int8(get_prefixq(src, dst), 0x33);
   emit_operand(dst, src, 0);
 }
 
+void Assembler::exorq(Register dst, Register src1, Address src2, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src2, dst->encoding(), src1->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0x33);
+  emit_operand(src1, src2, 0);
+}
+
 void Assembler::xorq(Register dst, int32_t imm32) {
   (void) prefixq_and_encode(dst->encoding());
   emit_arith(0x81, 0xF0, dst, imm32);
 }
 
+void Assembler::exorq(Register dst, Register src, int32_t imm32, bool no_flags) {
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  evex_prefix_and_encode_ndd(0, dst->encoding(), src->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_arith(0x81, 0xF0, src, imm32);
+}
+
 void Assembler::xorq(Address dst, int32_t imm32) {
   InstructionMark im(this);
   prefixq(dst);
   emit_arith_operand(0x81, as_Register(6), dst, imm32);
 }
 
+void Assembler::exorq(Register dst, Address src, int32_t imm32, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src, dst->encoding(), 0, VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_arith_operand(0x81, as_Register(6), src, imm32);
+}
+
 void Assembler::xorq(Address dst, Register src) {
   InstructionMark im(this);
   emit_prefix_and_int8(get_prefixq(dst, src), 0x31);
   emit_operand(src, dst, 0);
 }
 
+void Assembler::exorq(Register dst, Address src1, Register src2, bool no_flags) {
+  InstructionMark im(this);
+  InstructionAttr attributes(AVX_128bit, /* vex_w */ true, /* legacy_mode */ false, /* no_mask_reg */ true, /* uses_vl */ false);
+  attributes.set_address_attributes(/* tuple_type */ EVEX_NOSCALE, /* input_size_in_bits */ EVEX_64bit);
+  evex_prefix_ndd(src1, dst->encoding(), src2->encoding(), VEX_SIMD_NONE, VEX_OPCODE_0F_3C, &attributes, no_flags);
+  emit_int8(0x31);
+  emit_operand(src2, src1, 0);
+}
+
 #endif // !LP64
 
 void InstructionAttr::set_address_attributes(int tuple_type, int input_size_in_bits) {
diff --git a/src/hotspot/cpu/x86/assembler_x86.hpp b/src/hotspot/cpu/x86/assembler_x86.hpp
index 9d1a12ca8e5c8..9e7126417b06f 100644
--- a/src/hotspot/cpu/x86/assembler_x86.hpp
+++ b/src/hotspot/cpu/x86/assembler_x86.hpp
@@ -789,14 +789,26 @@ class Assembler : public AbstractAssembler  {
   void vex_prefix(bool vex_r, bool vex_b, bool vex_x, int nds_enc, VexSimdPrefix pre, VexOpcode opc);
 
   void evex_prefix(bool vex_r, bool vex_b, bool vex_x, bool evex_v, bool evex_r, bool evex_b,
-                       bool eevex_x, int nds_enc, VexSimdPrefix pre, VexOpcode opc);
+                       bool eevex_x, int nds_enc, VexSimdPrefix pre, VexOpcode opc, bool no_flags = false);
+
+  void evex_prefix_ndd(Address adr, int ndd_enc, int xreg_enc, VexSimdPrefix pre, VexOpcode opc,
+                       InstructionAttr *attributes, bool no_flags = false);
+
+  void evex_prefix_nf(Address adr, int ndd_enc, int xreg_enc, VexSimdPrefix pre, VexOpcode opc,
+                      InstructionAttr *attributes, bool no_flags = false);
 
   void vex_prefix(Address adr, int nds_enc, int xreg_enc, VexSimdPrefix pre, VexOpcode opc,
-                  InstructionAttr *attributes);
+                  InstructionAttr *attributes, bool nds_is_ndd = false, bool no_flags = false);
 
   int  vex_prefix_and_encode(int dst_enc, int nds_enc, int src_enc,
                              VexSimdPrefix pre, VexOpcode opc,
-                             InstructionAttr *attributes, bool src_is_gpr = false);
+                             InstructionAttr *attributes, bool src_is_gpr = false, bool nds_is_ndd = false, bool no_flags = false);
+
+  int  evex_prefix_and_encode_ndd(int dst_enc, int nds_enc, int src_enc, VexSimdPrefix pre, VexOpcode opc,
+                                  InstructionAttr *attributes, bool no_flags = false);
+
+  int  evex_prefix_and_encode_nf(int dst_enc, int nds_enc, int src_enc, VexSimdPrefix pre, VexOpcode opc,
+                                 InstructionAttr *attributes, bool no_flags = false);
 
   void simd_prefix(XMMRegister xreg, XMMRegister nds, Address adr, VexSimdPrefix pre,
                    VexOpcode opc, InstructionAttr *attributes);
@@ -941,13 +953,20 @@ class Assembler : public AbstractAssembler  {
   // the product flag UseIncDec value.
 
   void decl(Register dst);
+  void edecl(Register dst, Register src, bool no_flags);
   void decl(Address dst);
+  void edecl(Register dst, Address src, bool no_flags);
   void decq(Address dst);
+  void edecq(Register dst, Address src, bool no_flags);
 
   void incl(Register dst);
+  void eincl(Register dst, Register src, bool no_flags);
   void incl(Address dst);
+  void eincl(Register dst, Address src, bool no_flags);
   void incq(Register dst);
+  void eincq(Register dst, Register src, bool no_flags);
   void incq(Address dst);
+  void eincq(Register dst, Address src, bool no_flags);
 
   // New cpus require use of movsd and movss to avoid partial register stall
   // when loading from memory. But for old Opteron use movlpd instead of movsd.
@@ -1031,6 +1050,7 @@ class Assembler : public AbstractAssembler  {
 #endif
   void vzeroupper_uncached();
   void decq(Register dst);
+  void edecq(Register dst, Register src, bool no_flags);
 
   void pusha();
   void popa();
@@ -1072,23 +1092,35 @@ class Assembler : public AbstractAssembler  {
   void addw(Address dst, Register src);
 
   void addl(Address dst, int32_t imm32);
+  void eaddl(Register dst, Address src, int32_t imm32, bool no_flags);
   void addl(Address dst, Register src);
+  void eaddl(Register dst, Address src1, Register src2, bool no_flags);
   void addl(Register dst, int32_t imm32);
+  void eaddl(Register dst, Register src, int32_t imm32, bool no_flags);
   void addl(Register dst, Address src);
+  void eaddl(Register dst, Register src1, Address src2, bool no_flags);
   void addl(Register dst, Register src);
+  void eaddl(Register dst, Register src1, Register src2, bool no_flags);
 
   void addq(Address dst, int32_t imm32);
+  void eaddq(Register dst, Address src, int32_t imm32, bool no_flags);
   void addq(Address dst, Register src);
+  void eaddq(Register dst, Address src1, Register src2, bool no_flags);
   void addq(Register dst, int32_t imm32);
+  void eaddq(Register dst, Register src, int32_t imm32, bool no_flags);
   void addq(Register dst, Address src);
+  void eaddq(Register dst, Register src1, Address src2, bool no_flags);
   void addq(Register dst, Register src);
+  void eaddq(Register dst, Register src1, Register src2, bool no_flags);
 
 #ifdef _LP64
  //Add Unsigned Integers with Carry Flag
   void adcxq(Register dst, Register src);
+  void eadcxq(Register dst, Register src1, Register src2);
 
  //Add Unsigned Integers with Overflow Flag
   void adoxq(Register dst, Register src);
+  void eadoxq(Register dst, Register src1, Register src2);
 #endif
 
   void addr_nop_4();
@@ -1122,16 +1154,25 @@ class Assembler : public AbstractAssembler  {
   void andb(Address dst, Register src);
 
   void andl(Address  dst, int32_t imm32);
+  void eandl(Register dst, Address  src, int32_t imm32, bool no_flags);
   void andl(Register dst, int32_t imm32);
+  void eandl(Register dst, Register src, int32_t imm32, bool no_flags);
   void andl(Register dst, Address src);
+  void eandl(Register dst, Register src1, Address src2, bool no_flags);
   void andl(Register dst, Register src);
+  void eandl(Register dst, Register src1, Register src2, bool no_flags);
   void andl(Address dst, Register src);
 
   void andq(Address  dst, int32_t imm32);
+  void eandq(Register dst, Address  src, int32_t imm32, bool no_flags);
   void andq(Register dst, int32_t imm32);
+  void eandq(Register dst, Register src, int32_t imm32, bool no_flags);
   void andq(Register dst, Address src);
+  void eandq(Register dst, Register src1, Address src2, bool no_flags);
   void andq(Register dst, Register src);
+  void eandq(Register dst, Register src1, Register src2, bool no_flags);
   void andq(Address dst, Register src);
+  void eandq(Register dst, Address src1, Register src2, bool no_flags);
 
   // BMI instructions
   void andnl(Register dst, Register src1, Register src2);
@@ -1182,10 +1223,14 @@ class Assembler : public AbstractAssembler  {
   void clwb(Address adr);
 
   void cmovl(Condition cc, Register dst, Register src);
+  void ecmovl(Condition cc, Register dst, Register src1, Register src2);
   void cmovl(Condition cc, Register dst, Address src);
+  void ecmovl(Condition cc, Register dst, Register src1, Address src2);
 
   void cmovq(Condition cc, Register dst, Register src);
+  void ecmovq(Condition cc, Register dst, Register src1, Register src2);
   void cmovq(Condition cc, Register dst, Address src);
+  void ecmovq(Condition cc, Register dst, Register src1, Address src2);
 
 
   void cmpb(Address dst, int imm8);
@@ -1488,25 +1533,41 @@ class Assembler : public AbstractAssembler  {
   void hlt();
 
   void idivl(Register src);
+  void eidivl(Register src, bool no_flags);
   void divl(Register src); // Unsigned division
+  void edivl(Register src, bool no_flags); // Unsigned division
 
 #ifdef _LP64
   void idivq(Register src);
+  void eidivq(Register src, bool no_flags);
   void divq(Register src); // Unsigned division
+  void edivq(Register src, bool no_flags); // Unsigned division
 #endif
 
   void imull(Register src);
+  void eimull(Register src, bool no_flags);
   void imull(Register dst, Register src);
+  void eimull(Register dst, Register src1, Register src2, bool no_flags);
   void imull(Register dst, Register src, int value);
+  void eimull(Register dst, Register src, int value, bool no_flags);
   void imull(Register dst, Address src, int value);
+  void eimull(Register dst, Address src, int value, bool no_flags);
   void imull(Register dst, Address src);
+  void eimull(Register dst, Register src1, Address src2, bool no_flags);
 
 #ifdef _LP64
   void imulq(Register dst, Register src);
+  void eimulq(Register dst, Register src, bool no_flags);
+  void eimulq(Register dst, Register src1, Register src2, bool no_flags);
   void imulq(Register dst, Register src, int value);
+  void eimulq(Register dst, Register src, int value, bool no_flags);
   void imulq(Register dst, Address src, int value);
+  void eimulq(Register dst, Address src, int value, bool no_flags);
   void imulq(Register dst, Address src);
+  void eimulq(Register dst, Address src, bool no_flags);
+  void eimulq(Register dst, Register src1, Address src2, bool no_flags);
   void imulq(Register dst);
+  void eimulq(Register dst, bool no_flags);
 #endif
 
   // jcc is the generic conditional branch generator to run-
@@ -1565,11 +1626,15 @@ class Assembler : public AbstractAssembler  {
   void size_prefix();
 
   void lzcntl(Register dst, Register src);
+  void elzcntl(Register dst, Register src, bool no_flags);
   void lzcntl(Register dst, Address src);
+  void elzcntl(Register dst, Address src, bool no_flags);
 
 #ifdef _LP64
   void lzcntq(Register dst, Register src);
+  void elzcntq(Register dst, Register src, bool no_flags);
   void lzcntq(Register dst, Address src);
+  void elzcntq(Register dst, Address src, bool no_flags);
 #endif
 
   enum Membar_mask_bits {
@@ -1785,11 +1850,15 @@ class Assembler : public AbstractAssembler  {
 
   // Unsigned multiply with RAX destination register
   void mull(Address src);
+  void emull(Address src, bool no_flags);
   void mull(Register src);
+  void emull(Register src, bool no_flags);
 
 #ifdef _LP64
   void mulq(Address src);
+  void emulq(Address src, bool no_flags);
   void mulq(Register src);
+  void emulq(Register src, bool no_flags);
   void mulxq(Register dst1, Register dst2, Register src);
 #endif
 
@@ -1802,19 +1871,25 @@ class Assembler : public AbstractAssembler  {
   void mulss(XMMRegister dst, XMMRegister src);
 
   void negl(Register dst);
+  void enegl(Register dst, Register src, bool no_flags);
   void negl(Address dst);
+  void enegl(Register dst, Address src, bool no_flags);
 
 #ifdef _LP64
   void negq(Register dst);
+  void enegq(Register dst, Register src, bool no_flags);
   void negq(Address dst);
+  void enegq(Register dst, Address src, bool no_flags);
 #endif
 
   void nop(uint i = 1);
 
   void notl(Register dst);
+  void enotl(Register dst, Register src);
 
 #ifdef _LP64
   void notq(Register dst);
+  void enotq(Register dst, Register src);
 
   void btsq(Address dst, int imm8);
   void btrq(Address dst, int imm8);
@@ -1822,21 +1897,37 @@ class Assembler : public AbstractAssembler  {
 #endif
   void btq(Register dst, Register src);
 
+  void orw(Register dst, Register src);
+  void eorw(Register dst, Register src1, Register src2, bool no_flags);
+
   void orl(Address dst, int32_t imm32);
+  void eorl(Register dst, Address  src, int32_t imm32, bool no_flags);
   void orl(Register dst, int32_t imm32);
+  void eorl(Register dst, Register src, int32_t imm32, bool no_flags);
   void orl(Register dst, Address src);
+  void eorl(Register dst, Register src1, Address src2, bool no_flags);
   void orl(Register dst, Register src);
+  void eorl(Register dst, Register src1, Register src2, bool no_flags);
   void orl(Address dst, Register src);
+  void eorl(Register dst, Address src1, Register src2, bool no_flags);
 
   void orb(Address dst, int imm8);
+  void eorb(Register dst, Address  src, int imm8, bool no_flags);
   void orb(Address dst, Register src);
+  void eorb(Register dst, Address src1, Register src2, bool no_flags);
 
   void orq(Address dst, int32_t imm32);
+  void eorq(Register dst, Address  src, int32_t imm32, bool no_flags);
   void orq(Address dst, Register src);
+  void eorq(Register dst, Address src1, Register src2, bool no_flags);
   void orq(Register dst, int32_t imm32);
+  void eorq(Register dst, Register src, int32_t imm32, bool no_flags);
   void orq_imm32(Register dst, int32_t imm32);
+  void eorq_imm32(Register dst, Register src, int32_t imm32, bool no_flags);
   void orq(Register dst, Address src);
+  void eorq(Register dst, Register src1, Address src2, bool no_flags);
   void orq(Register dst, Register src);
+  void eorq(Register dst, Register src1, Register src2, bool no_flags);
 
   // Pack with signed saturation
   void packsswb(XMMRegister dst, XMMRegister src);
@@ -2022,7 +2113,9 @@ class Assembler : public AbstractAssembler  {
 #endif
 
   void popcntl(Register dst, Address src);
+  void epopcntl(Register dst, Address src, bool no_flags);
   void popcntl(Register dst, Register src);
+  void epopcntl(Register dst, Register src, bool no_flags);
 
   void evpopcntb(XMMRegister dst, KRegister mask, XMMRegister src, bool merge, int vector_len);
   void evpopcntw(XMMRegister dst, KRegister mask, XMMRegister src, bool merge, int vector_len);
@@ -2031,7 +2124,9 @@ class Assembler : public AbstractAssembler  {
 
 #ifdef _LP64
   void popcntq(Register dst, Address src);
+  void epopcntq(Register dst, Address src, bool no_flags);
   void popcntq(Register dst, Register src);
+  void epopcntq(Register dst, Register src, bool no_flags);
 #endif
 
   // Prefetches (SSE, SSE2, 3DNOW only)
@@ -2131,10 +2226,13 @@ class Assembler : public AbstractAssembler  {
   void pushq(Address src);
 
   void rcll(Register dst, int imm8);
+  void ercll(Register dst, Register src, int imm8);
 
   void rclq(Register dst, int imm8);
+  void erclq(Register dst, Register src, int imm8);
 
   void rcrq(Register dst, int imm8);
+  void ercrq(Register dst, Register src, int imm8);
 
   void rcpps(XMMRegister dst, XMMRegister src);
 
@@ -2145,18 +2243,26 @@ class Assembler : public AbstractAssembler  {
   void ret(int imm16);
 
   void roll(Register dst);
+  void eroll(Register dst, Register src, bool no_flags);
 
   void roll(Register dst, int imm8);
+  void eroll(Register dst, Register src, int imm8, bool no_flags);
 
   void rorl(Register dst);
+  void erorl(Register dst, Register src, bool no_flags);
 
   void rorl(Register dst, int imm8);
+  void erorl(Register dst, Register src, int imm8, bool no_flags);
 
 #ifdef _LP64
   void rolq(Register dst);
+  void erolq(Register dst, Register src, bool no_flags);
   void rolq(Register dst, int imm8);
+  void erolq(Register dst, Register src, int imm8, bool no_flags);
   void rorq(Register dst);
+  void erorq(Register dst, Register src, bool no_flags);
   void rorq(Register dst, int imm8);
+  void erorq(Register dst, Register src, int imm8, bool no_flags);
   void rorxl(Register dst, Register src, int imm8);
   void rorxl(Register dst, Address src, int imm8);
   void rorxq(Register dst, Register src, int imm8);
@@ -2166,25 +2272,41 @@ class Assembler : public AbstractAssembler  {
   void sahf();
 
   void sall(Register dst, int imm8);
+  void esall(Register dst, Register src, int imm8, bool no_flags);
   void sall(Register dst);
+  void esall(Register dst, Register src, bool no_flags);
   void sall(Address dst, int imm8);
+  void esall(Register dst, Address src, int imm8, bool no_flags);
   void sall(Address dst);
+  void esall(Register dst, Address src, bool no_flags);
 
   void sarl(Address dst, int imm8);
+  void esarl(Register dst, Address src, int imm8, bool no_flags);
   void sarl(Address dst);
+  void esarl(Register dst, Address src, bool no_flags);
   void sarl(Register dst, int imm8);
+  void esarl(Register dst, Register src, int imm8, bool no_flags);
   void sarl(Register dst);
+  void esarl(Register dst, Register src, bool no_flags);
 
 #ifdef _LP64
   void salq(Register dst, int imm8);
+  void esalq(Register dst, Register src, int imm8, bool no_flags);
   void salq(Register dst);
+  void esalq(Register dst, Register src, bool no_flags);
   void salq(Address dst, int imm8);
+  void esalq(Register dst, Address src, int imm8, bool no_flags);
   void salq(Address dst);
+  void esalq(Register dst, Address src, bool no_flags);
 
   void sarq(Address dst, int imm8);
+  void esarq(Register dst, Address src, int imm8, bool no_flags);
   void sarq(Address dst);
+  void esarq(Register dst, Address src, bool no_flags);
   void sarq(Register dst, int imm8);
+  void esarq(Register dst, Register src, int imm8, bool no_flags);
   void sarq(Register dst);
+  void esarq(Register dst, Register src, bool no_flags);
 #endif
 
   void sbbl(Address dst, int32_t imm32);
@@ -2216,29 +2338,47 @@ class Assembler : public AbstractAssembler  {
   void sha256msg2(XMMRegister dst, XMMRegister src);
 
   void shldl(Register dst, Register src);
+  void eshldl(Register dst, Register src1, Register src2, bool no_flags);
   void shldl(Register dst, Register src, int8_t imm8);
+  void eshldl(Register dst, Register src1, Register src2, int8_t imm8, bool no_flags);
   void shrdl(Register dst, Register src);
+  void eshrdl(Register dst, Register src1, Register src2, bool no_flags);
   void shrdl(Register dst, Register src, int8_t imm8);
+  void eshrdl(Register dst, Register src1, Register src2, int8_t imm8, bool no_flags);
 #ifdef _LP64
   void shldq(Register dst, Register src, int8_t imm8);
+  void eshldq(Register dst, Register src1, Register src2, int8_t imm8, bool no_flags);
   void shrdq(Register dst, Register src, int8_t imm8);
+  void eshrdq(Register dst, Register src1, Register src2, int8_t imm8, bool no_flags);
 #endif
 
   void shll(Register dst, int imm8);
+  void eshll(Register dst, Register src, int imm8, bool no_flags);
   void shll(Register dst);
+  void eshll(Register dst, Register src, bool no_flags);
 
   void shlq(Register dst, int imm8);
+  void eshlq(Register dst, Register src, int imm8, bool no_flags);
   void shlq(Register dst);
+  void eshlq(Register dst, Register src, bool no_flags);
 
   void shrl(Register dst, int imm8);
+  void eshrl(Register dst, Register src, int imm8, bool no_flags);
   void shrl(Register dst);
+  void eshrl(Register dst, Register src, bool no_flags);
   void shrl(Address dst);
+  void eshrl(Register dst, Address src, bool no_flags);
   void shrl(Address dst, int imm8);
+  void eshrl(Register dst, Address src, int imm8, bool no_flags);
 
   void shrq(Register dst, int imm8);
+  void eshrq(Register dst, Register src, int imm8, bool no_flags);
   void shrq(Register dst);
+  void eshrq(Register dst, Register src, bool no_flags);
   void shrq(Address dst);
+  void eshrq(Register dst, Address src, bool no_flags);
   void shrq(Address dst, int imm8);
+  void eshrq(Register dst, Address src, int imm8, bool no_flags);
 
   void smovl(); // QQQ generic?
 
@@ -2258,20 +2398,32 @@ class Assembler : public AbstractAssembler  {
   void stmxcsr( Address dst );
 
   void subl(Address dst, int32_t imm32);
+  void esubl(Register dst, Address src, int32_t imm32, bool no_flags);
   void subl(Address dst, Register src);
+  void esubl(Register dst, Address src1, Register src2, bool no_flags);
   void subl(Register dst, int32_t imm32);
+  void esubl(Register dst, Register src, int32_t imm32, bool no_flags);
   void subl(Register dst, Address src);
+  void esubl(Register dst, Register src1, Address src2, bool no_flags);
   void subl(Register dst, Register src);
+  void esubl(Register dst, Register src1, Register src2, bool no_flags);
 
   void subq(Address dst, int32_t imm32);
+  void esubq(Register dst, Address src, int32_t imm32, bool no_flags);
   void subq(Address dst, Register src);
+  void esubq(Register dst, Address src1, Register src2, bool no_flags);
   void subq(Register dst, int32_t imm32);
+  void esubq(Register dst, Register src, int32_t imm32, bool no_flags);
   void subq(Register dst, Address src);
+  void esubq(Register dst, Register src1, Address src2, bool no_flags);
   void subq(Register dst, Register src);
+  void esubq(Register dst, Register src1, Register src2, bool no_flags);
 
   // Force generation of a 4 byte immediate value even if it fits into 8bit
   void subl_imm32(Register dst, int32_t imm32);
+  void esubl_imm32(Register dst, Register src, int32_t imm32, bool no_flags);
   void subq_imm32(Register dst, int32_t imm32);
+  void esubq_imm32(Register dst, Register src, int32_t imm32, bool no_flags);
 
   // Subtract Scalar Double-Precision Floating-Point Values
   void subsd(XMMRegister dst, Address src);
@@ -2296,9 +2448,13 @@ class Assembler : public AbstractAssembler  {
 
   // BMI - count trailing zeros
   void tzcntl(Register dst, Register src);
+  void etzcntl(Register dst, Register src, bool no_flags);
   void tzcntl(Register dst, Address src);
+  void etzcntl(Register dst, Address src, bool no_flags);
   void tzcntq(Register dst, Register src);
+  void etzcntq(Register dst, Register src, bool no_flags);
   void tzcntq(Register dst, Address src);
+  void etzcntq(Register dst, Address src, bool no_flags);
 
   // Unordered Compare Scalar Double-Precision Floating-Point Values and set EFLAGS
   void ucomisd(XMMRegister dst, Address src);
@@ -2331,20 +2487,33 @@ class Assembler : public AbstractAssembler  {
   void xgetbv();
 
   void xorl(Register dst, int32_t imm32);
+  void exorl(Register dst, Register src, int32_t imm32, bool no_flags);
   void xorl(Address dst, int32_t imm32);
+  void exorl(Register dst, Address  src, int32_t imm32, bool no_flags);
   void xorl(Register dst, Address src);
+  void exorl(Register dst, Register src1, Address src2, bool no_flags);
   void xorl(Register dst, Register src);
+  void exorl(Register dst, Register src1, Register src2, bool no_flags);
   void xorl(Address dst, Register src);
+  void exorl(Register dst, Address src1, Register src2, bool no_flags);
 
   void xorb(Address dst, Register src);
+  void exorb(Register dst, Address src1, Register src2, bool no_flags);
   void xorb(Register dst, Address src);
+  void exorb(Register dst, Register src1, Address src2, bool no_flags);
   void xorw(Register dst, Address src);
+  void exorw(Register dst, Register src1, Address src2, bool no_flags);
 
   void xorq(Register dst, Address src);
+  void exorq(Register dst, Register src1, Address src2, bool no_flags);
   void xorq(Address dst, int32_t imm32);
+  void exorq(Register dst, Address  src, int32_t imm32, bool no_flags);
   void xorq(Register dst, Register src);
+  void exorq(Register dst, Register src1, Register src2, bool no_flags);
   void xorq(Register dst, int32_t imm32);
+  void exorq(Register dst, Register src, int32_t imm32, bool no_flags);
   void xorq(Address dst, Register src);
+  void exorq(Register dst, Address src1, Register src2, bool no_flags);
 
   // AVX 3-operands scalar instructions (encoded with VEX prefix)
 
diff --git a/src/hotspot/cpu/x86/c1_CodeStubs_x86.cpp b/src/hotspot/cpu/x86/c1_CodeStubs_x86.cpp
index 7d89b148ba22f..71ca9351f86c9 100644
--- a/src/hotspot/cpu/x86/c1_CodeStubs_x86.cpp
+++ b/src/hotspot/cpu/x86/c1_CodeStubs_x86.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999, 2023, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 1999, 2024, Oracle and/or its affiliates. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -110,7 +110,7 @@ void CounterOverflowStub::emit_code(LIR_Assembler* ce) {
   Metadata *m = _method->as_constant_ptr()->as_metadata();
   ce->store_parameter(m, 1);
   ce->store_parameter(_bci, 0);
-  __ call(RuntimeAddress(Runtime1::entry_for(Runtime1::counter_overflow_id)));
+  __ call(RuntimeAddress(Runtime1::entry_for(C1StubId::counter_overflow_id)));
   ce->add_call_info_here(_info);
   ce->verify_oop_map(_info);
   __ jmp(_continuation);
@@ -119,7 +119,7 @@ void CounterOverflowStub::emit_code(LIR_Assembler* ce) {
 void RangeCheckStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
   if (_info->deoptimize_on_exception()) {
-    address a = Runtime1::entry_for(Runtime1::predicate_failed_trap_id);
+    address a = Runtime1::entry_for(C1StubId::predicate_failed_trap_id);
     __ call(RuntimeAddress(a));
     ce->add_call_info_here(_info);
     ce->verify_oop_map(_info);
@@ -133,11 +133,11 @@ void RangeCheckStub::emit_code(LIR_Assembler* ce) {
   } else {
     ce->store_parameter(_index->as_jint(), 0);
   }
-  Runtime1::StubID stub_id;
+  C1StubId stub_id;
   if (_throw_index_out_of_bounds_exception) {
-    stub_id = Runtime1::throw_index_exception_id;
+    stub_id = C1StubId::throw_index_exception_id;
   } else {
-    stub_id = Runtime1::throw_range_check_failed_id;
+    stub_id = C1StubId::throw_range_check_failed_id;
     ce->store_parameter(_array->as_pointer_register(), 1);
   }
   __ call(RuntimeAddress(Runtime1::entry_for(stub_id)));
@@ -152,7 +152,7 @@ PredicateFailedStub::PredicateFailedStub(CodeEmitInfo* info) {
 
 void PredicateFailedStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
-  address a = Runtime1::entry_for(Runtime1::predicate_failed_trap_id);
+  address a = Runtime1::entry_for(C1StubId::predicate_failed_trap_id);
   __ call(RuntimeAddress(a));
   ce->add_call_info_here(_info);
   ce->verify_oop_map(_info);
@@ -164,7 +164,7 @@ void DivByZeroStub::emit_code(LIR_Assembler* ce) {
     ce->compilation()->implicit_exception_table()->append(_offset, __ offset());
   }
   __ bind(_entry);
-  __ call(RuntimeAddress(Runtime1::entry_for(Runtime1::throw_div0_exception_id)));
+  __ call(RuntimeAddress(Runtime1::entry_for(C1StubId::throw_div0_exception_id)));
   ce->add_call_info_here(_info);
   debug_only(__ should_not_reach_here());
 }
@@ -172,14 +172,14 @@ void DivByZeroStub::emit_code(LIR_Assembler* ce) {
 
 // Implementation of NewInstanceStub
 
-NewInstanceStub::NewInstanceStub(LIR_Opr klass_reg, LIR_Opr result, ciInstanceKlass* klass, CodeEmitInfo* info, Runtime1::StubID stub_id) {
+NewInstanceStub::NewInstanceStub(LIR_Opr klass_reg, LIR_Opr result, ciInstanceKlass* klass, CodeEmitInfo* info, C1StubId stub_id) {
   _result = result;
   _klass = klass;
   _klass_reg = klass_reg;
   _info = new CodeEmitInfo(info);
-  assert(stub_id == Runtime1::new_instance_id                 ||
-         stub_id == Runtime1::fast_new_instance_id            ||
-         stub_id == Runtime1::fast_new_instance_init_check_id,
+  assert(stub_id == C1StubId::new_instance_id                 ||
+         stub_id == C1StubId::fast_new_instance_id            ||
+         stub_id == C1StubId::fast_new_instance_init_check_id,
          "need new_instance id");
   _stub_id   = stub_id;
 }
@@ -212,7 +212,7 @@ void NewTypeArrayStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
   assert(_length->as_register() == rbx, "length must in rbx,");
   assert(_klass_reg->as_register() == rdx, "klass_reg must in rdx");
-  __ call(RuntimeAddress(Runtime1::entry_for(Runtime1::new_type_array_id)));
+  __ call(RuntimeAddress(Runtime1::entry_for(C1StubId::new_type_array_id)));
   ce->add_call_info_here(_info);
   ce->verify_oop_map(_info);
   assert(_result->as_register() == rax, "result must in rax,");
@@ -235,7 +235,7 @@ void NewObjectArrayStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
   assert(_length->as_register() == rbx, "length must in rbx,");
   assert(_klass_reg->as_register() == rdx, "klass_reg must in rdx");
-  __ call(RuntimeAddress(Runtime1::entry_for(Runtime1::new_object_array_id)));
+  __ call(RuntimeAddress(Runtime1::entry_for(C1StubId::new_object_array_id)));
   ce->add_call_info_here(_info);
   ce->verify_oop_map(_info);
   assert(_result->as_register() == rax, "result must in rax,");
@@ -247,11 +247,11 @@ void MonitorEnterStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
   ce->store_parameter(_obj_reg->as_register(),  1);
   ce->store_parameter(_lock_reg->as_register(), 0);
-  Runtime1::StubID enter_id;
+  C1StubId enter_id;
   if (ce->compilation()->has_fpu_code()) {
-    enter_id = Runtime1::monitorenter_id;
+    enter_id = C1StubId::monitorenter_id;
   } else {
-    enter_id = Runtime1::monitorenter_nofpu_id;
+    enter_id = C1StubId::monitorenter_nofpu_id;
   }
   __ call(RuntimeAddress(Runtime1::entry_for(enter_id)));
   ce->add_call_info_here(_info);
@@ -268,11 +268,11 @@ void MonitorExitStub::emit_code(LIR_Assembler* ce) {
   }
   ce->store_parameter(_lock_reg->as_register(), 0);
   // note: non-blocking leaf routine => no call info needed
-  Runtime1::StubID exit_id;
+  C1StubId exit_id;
   if (ce->compilation()->has_fpu_code()) {
-    exit_id = Runtime1::monitorexit_id;
+    exit_id = C1StubId::monitorexit_id;
   } else {
-    exit_id = Runtime1::monitorexit_nofpu_id;
+    exit_id = C1StubId::monitorexit_nofpu_id;
   }
   __ call(RuntimeAddress(Runtime1::entry_for(exit_id)));
   __ jmp(_continuation);
@@ -407,10 +407,10 @@ void PatchingStub::emit_code(LIR_Assembler* ce) {
   address target = nullptr;
   relocInfo::relocType reloc_type = relocInfo::none;
   switch (_id) {
-    case access_field_id:  target = Runtime1::entry_for(Runtime1::access_field_patching_id); break;
-    case load_klass_id:    target = Runtime1::entry_for(Runtime1::load_klass_patching_id); reloc_type = relocInfo::metadata_type; break;
-    case load_mirror_id:   target = Runtime1::entry_for(Runtime1::load_mirror_patching_id); reloc_type = relocInfo::oop_type; break;
-    case load_appendix_id:      target = Runtime1::entry_for(Runtime1::load_appendix_patching_id); reloc_type = relocInfo::oop_type; break;
+    case access_field_id:  target = Runtime1::entry_for(C1StubId::access_field_patching_id); break;
+    case load_klass_id:    target = Runtime1::entry_for(C1StubId::load_klass_patching_id); reloc_type = relocInfo::metadata_type; break;
+    case load_mirror_id:   target = Runtime1::entry_for(C1StubId::load_mirror_patching_id); reloc_type = relocInfo::oop_type; break;
+    case load_appendix_id:      target = Runtime1::entry_for(C1StubId::load_appendix_patching_id); reloc_type = relocInfo::oop_type; break;
     default: ShouldNotReachHere();
   }
   __ bind(call_patch);
@@ -440,7 +440,7 @@ void PatchingStub::emit_code(LIR_Assembler* ce) {
 void DeoptimizeStub::emit_code(LIR_Assembler* ce) {
   __ bind(_entry);
   ce->store_parameter(_trap_request, 0);
-  __ call(RuntimeAddress(Runtime1::entry_for(Runtime1::deoptimize_id)));
+  __ call(RuntimeAddress(Runtime1::entry_for(C1StubId::deoptimize_id)));
   ce->add_call_info_here(_info);
   DEBUG_ONLY(__ should_not_reach_here());
 }
@@ -450,9 +450,9 @@ void ImplicitNullCheckStub::emit_code(LIR_Assembler* ce) {
   address a;
   if (_info->deoptimize_on_exception()) {
     // Deoptimize, do not throw the exception, because it is probably wrong to do it here.
-    a = Runtime1::entry_for(Runtime1::predicate_failed_trap_id);
+    a = Runtime1::entry_for(C1StubId::predicate_failed_trap_id);
   } else {
-    a = Runtime1::entry_for(Runtime1::throw_null_pointer_exception_id);
+    a = Runtime1::entry_for(C1StubId::throw_null_pointer_exception_id);
   }
 
   ce->compilation()->implicit_exception_table()->append(_offset, __ offset());
diff --git a/src/hotspot/cpu/x86/c1_LIRAssembler_x86.cpp b/src/hotspot/cpu/x86/c1_LIRAssembler_x86.cpp
index e2fde10b98d86..c3444d5a5abce 100644
--- a/src/hotspot/cpu/x86/c1_LIRAssembler_x86.cpp
+++ b/src/hotspot/cpu/x86/c1_LIRAssembler_x86.cpp
@@ -399,7 +399,7 @@ int LIR_Assembler::emit_exception_handler() {
   __ verify_not_null_oop(rax);
 
   // search an exception handler (rax: exception oop, rdx: throwing pc)
-  __ call(RuntimeAddress(Runtime1::entry_for(Runtime1::handle_exception_from_callee_id)));
+  __ call(RuntimeAddress(Runtime1::entry_for(C1StubId::handle_exception_from_callee_id)));
   __ should_not_reach_here();
   guarantee(code_offset() - offset <= exception_handler_size(), "overflow");
   __ end_a_stub();
@@ -463,7 +463,7 @@ int LIR_Assembler::emit_unwind_handler() {
 
   // remove the activation and dispatch to the unwind handler
   __ remove_frame(initial_frame_size_in_bytes());
-  __ jump(RuntimeAddress(Runtime1::entry_for(Runtime1::unwind_exception_id)));
+  __ jump(RuntimeAddress(Runtime1::entry_for(C1StubId::unwind_exception_id)));
 
   // Emit the slow path assembly
   if (stub != nullptr) {
@@ -1566,7 +1566,7 @@ void LIR_Assembler::emit_opConvert(LIR_OpConvert* op) {
 
       // instruction sequence too long to inline it here
       {
-        __ call(RuntimeAddress(Runtime1::entry_for(Runtime1::fpu2long_stub_id)));
+        __ call(RuntimeAddress(Runtime1::entry_for(C1StubId::fpu2long_stub_id)));
       }
       break;
 #endif // _LP64
@@ -1781,7 +1781,7 @@ void LIR_Assembler::emit_typecheck_helper(LIR_OpTypeCheck *op, Label* success, L
 #else
         __ pushklass(k->constant_encoding(), noreg);
 #endif // _LP64
-        __ call(RuntimeAddress(Runtime1::entry_for(Runtime1::slow_subtype_check_id)));
+        __ call(RuntimeAddress(Runtime1::entry_for(C1StubId::slow_subtype_check_id)));
         __ pop(klass_RInfo);
         __ pop(klass_RInfo);
         // result is a boolean
@@ -1795,7 +1795,7 @@ void LIR_Assembler::emit_typecheck_helper(LIR_OpTypeCheck *op, Label* success, L
       // call out-of-line instance of __ check_klass_subtype_slow_path(...):
       __ push(klass_RInfo);
       __ push(k_RInfo);
-      __ call(RuntimeAddress(Runtime1::entry_for(Runtime1::slow_subtype_check_id)));
+      __ call(RuntimeAddress(Runtime1::entry_for(C1StubId::slow_subtype_check_id)));
       __ pop(klass_RInfo);
       __ pop(k_RInfo);
       // result is a boolean
@@ -1874,7 +1874,7 @@ void LIR_Assembler::emit_opTypeCheck(LIR_OpTypeCheck* op) {
     // call out-of-line instance of __ check_klass_subtype_slow_path(...):
     __ push(klass_RInfo);
     __ push(k_RInfo);
-    __ call(RuntimeAddress(Runtime1::entry_for(Runtime1::slow_subtype_check_id)));
+    __ call(RuntimeAddress(Runtime1::entry_for(C1StubId::slow_subtype_check_id)));
     __ pop(klass_RInfo);
     __ pop(k_RInfo);
     // result is a boolean
@@ -2893,7 +2893,7 @@ void LIR_Assembler::throw_op(LIR_Opr exceptionPC, LIR_Opr exceptionOop, CodeEmit
   // exception object is not added to oop map by LinearScan
   // (LinearScan assumes that no oops are in fixed registers)
   info->add_register_oop(exceptionOop);
-  Runtime1::StubID unwind_id;
+  C1StubId unwind_id;
 
   // get current pc information
   // pc is only needed if the method has an exception handler, the unwind code does not need it.
@@ -2905,9 +2905,9 @@ void LIR_Assembler::throw_op(LIR_Opr exceptionPC, LIR_Opr exceptionOop, CodeEmit
   __ verify_not_null_oop(rax);
   // search an exception handler (rax: exception oop, rdx: throwing pc)
   if (compilation()->has_fpu_code()) {
-    unwind_id = Runtime1::handle_exception_id;
+    unwind_id = C1StubId::handle_exception_id;
   } else {
-    unwind_id = Runtime1::handle_exception_nofpu_id;
+    unwind_id = C1StubId::handle_exception_nofpu_id;
   }
   __ call(RuntimeAddress(Runtime1::entry_for(unwind_id)));
 
@@ -3262,7 +3262,7 @@ void LIR_Assembler::emit_arraycopy(LIR_OpArrayCopy* op) {
 
       __ push(src);
       __ push(dst);
-      __ call(RuntimeAddress(Runtime1::entry_for(Runtime1::slow_subtype_check_id)));
+      __ call(RuntimeAddress(Runtime1::entry_for(C1StubId::slow_subtype_check_id)));
       __ pop(dst);
       __ pop(src);
 
diff --git a/src/hotspot/cpu/x86/c1_LIRGenerator_x86.cpp b/src/hotspot/cpu/x86/c1_LIRGenerator_x86.cpp
index d3add6975b4f2..ff237d16d2216 100644
--- a/src/hotspot/cpu/x86/c1_LIRGenerator_x86.cpp
+++ b/src/hotspot/cpu/x86/c1_LIRGenerator_x86.cpp
@@ -1430,7 +1430,7 @@ void LIRGenerator::do_NewMultiArray(NewMultiArray* x) {
   args->append(rank);
   args->append(varargs);
   LIR_Opr reg = result_register_for(x->type());
-  __ call_runtime(Runtime1::entry_for(Runtime1::new_multi_array_id),
+  __ call_runtime(Runtime1::entry_for(C1StubId::new_multi_array_id),
                   LIR_OprFact::illegalOpr,
                   reg, args, info);
 
@@ -1463,12 +1463,12 @@ void LIRGenerator::do_CheckCast(CheckCast* x) {
   CodeStub* stub;
   if (x->is_incompatible_class_change_check()) {
     assert(patching_info == nullptr, "can't patch this");
-    stub = new SimpleExceptionStub(Runtime1::throw_incompatible_class_change_error_id, LIR_OprFact::illegalOpr, info_for_exception);
+    stub = new SimpleExceptionStub(C1StubId::throw_incompatible_class_change_error_id, LIR_OprFact::illegalOpr, info_for_exception);
   } else if (x->is_invokespecial_receiver_check()) {
     assert(patching_info == nullptr, "can't patch this");
     stub = new DeoptimizeStub(info_for_exception, Deoptimization::Reason_class_check, Deoptimization::Action_none);
   } else {
-    stub = new SimpleExceptionStub(Runtime1::throw_class_cast_exception_id, obj.result(), info_for_exception);
+    stub = new SimpleExceptionStub(C1StubId::throw_class_cast_exception_id, obj.result(), info_for_exception);
   }
   LIR_Opr reg = rlock_result(x);
   LIR_Opr tmp3 = LIR_OprFact::illegalOpr;
diff --git a/src/hotspot/cpu/x86/c1_MacroAssembler_x86.cpp b/src/hotspot/cpu/x86/c1_MacroAssembler_x86.cpp
index 4dcacd00a6339..bf5b90db5fcb0 100644
--- a/src/hotspot/cpu/x86/c1_MacroAssembler_x86.cpp
+++ b/src/hotspot/cpu/x86/c1_MacroAssembler_x86.cpp
@@ -271,7 +271,7 @@ void C1_MacroAssembler::initialize_object(Register obj, Register klass, Register
 
   if (CURRENT_ENV->dtrace_alloc_probes()) {
     assert(obj == rax, "must be");
-    call(RuntimeAddress(Runtime1::entry_for(Runtime1::dtrace_object_alloc_id)));
+    call(RuntimeAddress(Runtime1::entry_for(C1StubId::dtrace_object_alloc_id)));
   }
 
   verify_oop(obj);
@@ -309,7 +309,7 @@ void C1_MacroAssembler::allocate_array(Register obj, Register len, Register t1,
 
   if (CURRENT_ENV->dtrace_alloc_probes()) {
     assert(obj == rax, "must be");
-    call(RuntimeAddress(Runtime1::entry_for(Runtime1::dtrace_object_alloc_id)));
+    call(RuntimeAddress(Runtime1::entry_for(C1StubId::dtrace_object_alloc_id)));
   }
 
   verify_oop(obj);
diff --git a/src/hotspot/cpu/x86/c1_Runtime1_x86.cpp b/src/hotspot/cpu/x86/c1_Runtime1_x86.cpp
index 11b39ce15eb1a..1ccb06df48937 100644
--- a/src/hotspot/cpu/x86/c1_Runtime1_x86.cpp
+++ b/src/hotspot/cpu/x86/c1_Runtime1_x86.cpp
@@ -60,7 +60,7 @@ int StubAssembler::call_RT(Register oop_result1, Register metadata_result, addre
 #ifdef _LP64
   // At a method handle call, the stack may not be properly aligned
   // when returning with an exception.
-  align_stack = (stub_id() == Runtime1::handle_exception_from_callee_id);
+  align_stack = (stub_id() == (int)C1StubId::handle_exception_from_callee_id);
 #endif
 
 #ifdef _LP64
@@ -124,10 +124,10 @@ int StubAssembler::call_RT(Register oop_result1, Register metadata_result, addre
     if (frame_size() == no_frame_size) {
       leave();
       jump(RuntimeAddress(StubRoutines::forward_exception_entry()));
-    } else if (_stub_id == Runtime1::forward_exception_id) {
+    } else if (_stub_id == (int)C1StubId::forward_exception_id) {
       should_not_reach_here();
     } else {
-      jump(RuntimeAddress(Runtime1::entry_for(Runtime1::forward_exception_id)));
+      jump(RuntimeAddress(Runtime1::entry_for(C1StubId::forward_exception_id)));
     }
     bind(L);
   }
@@ -671,7 +671,7 @@ OopMapSet* Runtime1::generate_exception_throw(StubAssembler* sasm, address targe
 }
 
 
-OopMapSet* Runtime1::generate_handle_exception(StubID id, StubAssembler *sasm) {
+OopMapSet* Runtime1::generate_handle_exception(C1StubId id, StubAssembler *sasm) {
   __ block_comment("generate_handle_exception");
 
   // incoming parameters
@@ -684,7 +684,7 @@ OopMapSet* Runtime1::generate_handle_exception(StubID id, StubAssembler *sasm) {
   OopMapSet* oop_maps = new OopMapSet();
   OopMap* oop_map = nullptr;
   switch (id) {
-  case forward_exception_id:
+  case C1StubId::forward_exception_id:
     // We're handling an exception in the context of a compiled frame.
     // The registers have been saved in the standard places.  Perform
     // an exception lookup in the caller and dispatch to the handler
@@ -703,12 +703,12 @@ OopMapSet* Runtime1::generate_handle_exception(StubID id, StubAssembler *sasm) {
     __ movptr(Address(thread, JavaThread::vm_result_offset()),   NULL_WORD);
     __ movptr(Address(thread, JavaThread::vm_result_2_offset()), NULL_WORD);
     break;
-  case handle_exception_nofpu_id:
-  case handle_exception_id:
+  case C1StubId::handle_exception_nofpu_id:
+  case C1StubId::handle_exception_id:
     // At this point all registers MAY be live.
-    oop_map = save_live_registers(sasm, 1 /*thread*/, id != handle_exception_nofpu_id);
+    oop_map = save_live_registers(sasm, 1 /*thread*/, id != C1StubId::handle_exception_nofpu_id);
     break;
-  case handle_exception_from_callee_id: {
+  case C1StubId::handle_exception_from_callee_id: {
     // At this point all registers except exception oop (RAX) and
     // exception pc (RDX) are dead.
     const int frame_size = 2 /*BP, return address*/ NOT_LP64(+ 1 /*thread*/) WIN64_ONLY(+ frame::arg_reg_save_area_bytes / BytesPerWord);
@@ -775,13 +775,13 @@ OopMapSet* Runtime1::generate_handle_exception(StubID id, StubAssembler *sasm) {
   __ movptr(Address(rbp, 1*BytesPerWord), rax);
 
   switch (id) {
-  case forward_exception_id:
-  case handle_exception_nofpu_id:
-  case handle_exception_id:
+  case C1StubId::forward_exception_id:
+  case C1StubId::handle_exception_nofpu_id:
+  case C1StubId::handle_exception_id:
     // Restore the registers that were saved at the beginning.
-    restore_live_registers(sasm, id != handle_exception_nofpu_id);
+    restore_live_registers(sasm, id != C1StubId::handle_exception_nofpu_id);
     break;
-  case handle_exception_from_callee_id:
+  case C1StubId::handle_exception_from_callee_id:
     // WIN64_ONLY: No need to add frame::arg_reg_save_area_bytes to SP
     // since we do a leave anyway.
 
@@ -935,7 +935,7 @@ OopMapSet* Runtime1::generate_patching(StubAssembler* sasm, address target) {
 
     __ testptr(rax, rax);                                   // have we deoptimized?
     __ jump_cc(Assembler::equal,
-               RuntimeAddress(Runtime1::entry_for(Runtime1::forward_exception_id)));
+               RuntimeAddress(Runtime1::entry_for(C1StubId::forward_exception_id)));
 
     // the deopt blob expects exceptions in the special fields of
     // JavaThread, so copy and clear pending exception.
@@ -1007,7 +1007,7 @@ OopMapSet* Runtime1::generate_patching(StubAssembler* sasm, address target) {
 }
 
 
-OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
+OopMapSet* Runtime1::generate_code_for(C1StubId id, StubAssembler* sasm) {
 
   // for better readability
   const bool must_gc_arguments = true;
@@ -1019,7 +1019,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
   // stub code & info for the different stubs
   OopMapSet* oop_maps = nullptr;
   switch (id) {
-    case forward_exception_id:
+    case C1StubId::forward_exception_id:
       {
         oop_maps = generate_handle_exception(id, sasm);
         __ leave();
@@ -1027,19 +1027,19 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case new_instance_id:
-    case fast_new_instance_id:
-    case fast_new_instance_init_check_id:
+    case C1StubId::new_instance_id:
+    case C1StubId::fast_new_instance_id:
+    case C1StubId::fast_new_instance_init_check_id:
       {
         Register klass = rdx; // Incoming
         Register obj   = rax; // Result
 
-        if (id == new_instance_id) {
+        if (id == C1StubId::new_instance_id) {
           __ set_info("new_instance", dont_gc_arguments);
-        } else if (id == fast_new_instance_id) {
+        } else if (id == C1StubId::fast_new_instance_id) {
           __ set_info("fast new_instance", dont_gc_arguments);
         } else {
-          assert(id == fast_new_instance_init_check_id, "bad StubID");
+          assert(id == C1StubId::fast_new_instance_init_check_id, "bad C1StubId");
           __ set_info("fast new_instance init check", dont_gc_arguments);
         }
 
@@ -1058,7 +1058,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
 
       break;
 
-    case counter_overflow_id:
+    case C1StubId::counter_overflow_id:
       {
         Register bci = rax, method = rbx;
         __ enter();
@@ -1076,14 +1076,14 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case new_type_array_id:
-    case new_object_array_id:
+    case C1StubId::new_type_array_id:
+    case C1StubId::new_object_array_id:
       {
         Register length   = rbx; // Incoming
         Register klass    = rdx; // Incoming
         Register obj      = rax; // Result
 
-        if (id == new_type_array_id) {
+        if (id == C1StubId::new_type_array_id) {
           __ set_info("new_type_array", dont_gc_arguments);
         } else {
           __ set_info("new_object_array", dont_gc_arguments);
@@ -1096,7 +1096,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
           Register t0 = obj;
           __ movl(t0, Address(klass, Klass::layout_helper_offset()));
           __ sarl(t0, Klass::_lh_array_tag_shift);
-          int tag = ((id == new_type_array_id)
+          int tag = ((id == C1StubId::new_type_array_id)
                      ? Klass::_lh_array_tag_type_value
                      : Klass::_lh_array_tag_obj_value);
           __ cmpl(t0, tag);
@@ -1110,7 +1110,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
         __ enter();
         OopMap* map = save_live_registers(sasm, 3);
         int call_offset;
-        if (id == new_type_array_id) {
+        if (id == C1StubId::new_type_array_id) {
           call_offset = __ call_RT(obj, noreg, CAST_FROM_FN_PTR(address, new_type_array), klass, length);
         } else {
           call_offset = __ call_RT(obj, noreg, CAST_FROM_FN_PTR(address, new_object_array), klass, length);
@@ -1128,7 +1128,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case new_multi_array_id:
+    case C1StubId::new_multi_array_id:
       { StubFrame f(sasm, "new_multi_array", dont_gc_arguments);
         // rax,: klass
         // rbx,: rank
@@ -1145,7 +1145,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case register_finalizer_id:
+    case C1StubId::register_finalizer_id:
       {
         __ set_info("register_finalizer", dont_gc_arguments);
 
@@ -1185,44 +1185,44 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case throw_range_check_failed_id:
+    case C1StubId::throw_range_check_failed_id:
       { StubFrame f(sasm, "range_check_failed", dont_gc_arguments);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_range_check_exception), true);
       }
       break;
 
-    case throw_index_exception_id:
+    case C1StubId::throw_index_exception_id:
       { StubFrame f(sasm, "index_range_check_failed", dont_gc_arguments);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_index_exception), true);
       }
       break;
 
-    case throw_div0_exception_id:
+    case C1StubId::throw_div0_exception_id:
       { StubFrame f(sasm, "throw_div0_exception", dont_gc_arguments);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_div0_exception), false);
       }
       break;
 
-    case throw_null_pointer_exception_id:
+    case C1StubId::throw_null_pointer_exception_id:
       { StubFrame f(sasm, "throw_null_pointer_exception", dont_gc_arguments);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_null_pointer_exception), false);
       }
       break;
 
-    case handle_exception_nofpu_id:
-    case handle_exception_id:
+    case C1StubId::handle_exception_nofpu_id:
+    case C1StubId::handle_exception_id:
       { StubFrame f(sasm, "handle_exception", dont_gc_arguments);
         oop_maps = generate_handle_exception(id, sasm);
       }
       break;
 
-    case handle_exception_from_callee_id:
+    case C1StubId::handle_exception_from_callee_id:
       { StubFrame f(sasm, "handle_exception_from_callee", dont_gc_arguments);
         oop_maps = generate_handle_exception(id, sasm);
       }
       break;
 
-    case unwind_exception_id:
+    case C1StubId::unwind_exception_id:
       { __ set_info("unwind_exception", dont_gc_arguments);
         // note: no stubframe since we are about to leave the current
         //       activation and we are calling a leaf VM function only.
@@ -1230,7 +1230,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case throw_array_store_exception_id:
+    case C1StubId::throw_array_store_exception_id:
       { StubFrame f(sasm, "throw_array_store_exception", dont_gc_arguments);
         // tos + 0: link
         //     + 1: return address
@@ -1238,19 +1238,19 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case throw_class_cast_exception_id:
+    case C1StubId::throw_class_cast_exception_id:
       { StubFrame f(sasm, "throw_class_cast_exception", dont_gc_arguments);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_class_cast_exception), true);
       }
       break;
 
-    case throw_incompatible_class_change_error_id:
+    case C1StubId::throw_incompatible_class_change_error_id:
       { StubFrame f(sasm, "throw_incompatible_class_cast_exception", dont_gc_arguments);
         oop_maps = generate_exception_throw(sasm, CAST_FROM_FN_PTR(address, throw_incompatible_class_change_error), false);
       }
       break;
 
-    case slow_subtype_check_id:
+    case C1StubId::slow_subtype_check_id:
       {
         // Typical calling sequence:
         // __ push(klass_RInfo);  // object klass or other subclass
@@ -1303,10 +1303,10 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case monitorenter_nofpu_id:
+    case C1StubId::monitorenter_nofpu_id:
       save_fpu_registers = false;
       // fall through
-    case monitorenter_id:
+    case C1StubId::monitorenter_id:
       {
         StubFrame f(sasm, "monitorenter", dont_gc_arguments);
         OopMap* map = save_live_registers(sasm, 3, save_fpu_registers);
@@ -1324,10 +1324,10 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case monitorexit_nofpu_id:
+    case C1StubId::monitorexit_nofpu_id:
       save_fpu_registers = false;
       // fall through
-    case monitorexit_id:
+    case C1StubId::monitorexit_id:
       {
         StubFrame f(sasm, "monitorexit", dont_gc_arguments);
         OopMap* map = save_live_registers(sasm, 2, save_fpu_registers);
@@ -1347,7 +1347,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case deoptimize_id:
+    case C1StubId::deoptimize_id:
       {
         StubFrame f(sasm, "deoptimize", dont_gc_arguments);
         const int num_rt_args = 2;  // thread, trap_request
@@ -1364,35 +1364,35 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case access_field_patching_id:
+    case C1StubId::access_field_patching_id:
       { StubFrame f(sasm, "access_field_patching", dont_gc_arguments);
         // we should set up register map
         oop_maps = generate_patching(sasm, CAST_FROM_FN_PTR(address, access_field_patching));
       }
       break;
 
-    case load_klass_patching_id:
+    case C1StubId::load_klass_patching_id:
       { StubFrame f(sasm, "load_klass_patching", dont_gc_arguments);
         // we should set up register map
         oop_maps = generate_patching(sasm, CAST_FROM_FN_PTR(address, move_klass_patching));
       }
       break;
 
-    case load_mirror_patching_id:
+    case C1StubId::load_mirror_patching_id:
       { StubFrame f(sasm, "load_mirror_patching", dont_gc_arguments);
         // we should set up register map
         oop_maps = generate_patching(sasm, CAST_FROM_FN_PTR(address, move_mirror_patching));
       }
       break;
 
-    case load_appendix_patching_id:
+    case C1StubId::load_appendix_patching_id:
       { StubFrame f(sasm, "load_appendix_patching", dont_gc_arguments);
         // we should set up register map
         oop_maps = generate_patching(sasm, CAST_FROM_FN_PTR(address, move_appendix_patching));
       }
       break;
 
-    case dtrace_object_alloc_id:
+    case C1StubId::dtrace_object_alloc_id:
       { // rax,: object
         StubFrame f(sasm, "dtrace_object_alloc", dont_gc_arguments);
         // we can't gc here so skip the oopmap but make sure that all
@@ -1407,7 +1407,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case fpu2long_stub_id:
+    case C1StubId::fpu2long_stub_id:
       {
 #ifdef _LP64
         Label done;
@@ -1496,7 +1496,7 @@ OopMapSet* Runtime1::generate_code_for(StubID id, StubAssembler* sasm) {
       }
       break;
 
-    case predicate_failed_trap_id:
+    case C1StubId::predicate_failed_trap_id:
       {
         StubFrame f(sasm, "predicate_failed_trap", dont_gc_arguments);
 
diff --git a/src/hotspot/share/c1/c1_CodeStubs.hpp b/src/hotspot/share/c1/c1_CodeStubs.hpp
index 04e379842e152..9abfa45785bda 100644
--- a/src/hotspot/share/c1/c1_CodeStubs.hpp
+++ b/src/hotspot/share/c1/c1_CodeStubs.hpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999, 2023, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 1999, 2024, Oracle and/or its affiliates. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -264,10 +264,10 @@ class NewInstanceStub: public CodeStub {
   LIR_Opr          _klass_reg;
   LIR_Opr          _result;
   CodeEmitInfo*    _info;
-  Runtime1::StubID _stub_id;
+  C1StubId         _stub_id;
 
  public:
-  NewInstanceStub(LIR_Opr klass_reg, LIR_Opr result, ciInstanceKlass* klass, CodeEmitInfo* info, Runtime1::StubID stub_id);
+  NewInstanceStub(LIR_Opr klass_reg, LIR_Opr result, ciInstanceKlass* klass, CodeEmitInfo* info, C1StubId stub_id);
   virtual void emit_code(LIR_Assembler* e);
   virtual CodeEmitInfo* info() const             { return _info; }
   virtual void visit(LIR_OpVisitState* visitor) {
@@ -515,11 +515,11 @@ class DeoptimizeStub : public CodeStub {
 class SimpleExceptionStub: public CodeStub {
  private:
   LIR_Opr          _obj;
-  Runtime1::StubID _stub;
+  C1StubId         _stub;
   CodeEmitInfo*    _info;
 
  public:
-  SimpleExceptionStub(Runtime1::StubID stub, LIR_Opr obj, CodeEmitInfo* info):
+  SimpleExceptionStub(C1StubId stub, LIR_Opr obj, CodeEmitInfo* info):
     _obj(obj), _stub(stub), _info(info) {
     FrameMap* f = Compilation::current()->frame_map();
     f->update_reserved_argument_area_size(2 * BytesPerWord);
@@ -546,7 +546,7 @@ class SimpleExceptionStub: public CodeStub {
 
 class ArrayStoreExceptionStub: public SimpleExceptionStub {
  public:
-  ArrayStoreExceptionStub(LIR_Opr obj, CodeEmitInfo* info): SimpleExceptionStub(Runtime1::throw_array_store_exception_id, obj, info) {}
+  ArrayStoreExceptionStub(LIR_Opr obj, CodeEmitInfo* info): SimpleExceptionStub(C1StubId::throw_array_store_exception_id, obj, info) {}
 #ifndef PRODUCT
   virtual void print_name(outputStream* out) const { out->print("ArrayStoreExceptionStub"); }
 #endif // PRODUCT
diff --git a/src/hotspot/share/c1/c1_LIRGenerator.cpp b/src/hotspot/share/c1/c1_LIRGenerator.cpp
index 7b519804bfecd..4e63736503fe0 100644
--- a/src/hotspot/share/c1/c1_LIRGenerator.cpp
+++ b/src/hotspot/share/c1/c1_LIRGenerator.cpp
@@ -659,7 +659,7 @@ void LIRGenerator::new_instance(LIR_Opr dst, ciInstanceKlass* klass, bool is_unr
   if (UseFastNewInstance && klass->is_loaded()
       && !Klass::layout_helper_needs_slow_path(klass->layout_helper())) {
 
-    Runtime1::StubID stub_id = klass->is_initialized() ? Runtime1::fast_new_instance_id : Runtime1::fast_new_instance_init_check_id;
+    C1StubId stub_id = klass->is_initialized() ? C1StubId::fast_new_instance_id : C1StubId::fast_new_instance_init_check_id;
 
     CodeStub* slow_path = new NewInstanceStub(klass_reg, dst, klass, info, stub_id);
 
@@ -670,7 +670,7 @@ void LIRGenerator::new_instance(LIR_Opr dst, ciInstanceKlass* klass, bool is_unr
     __ allocate_object(dst, scratch1, scratch2, scratch3, scratch4,
                        oopDesc::header_size(), instance_size, klass_reg, !klass->is_initialized(), slow_path);
   } else {
-    CodeStub* slow_path = new NewInstanceStub(klass_reg, dst, klass, info, Runtime1::new_instance_id);
+    CodeStub* slow_path = new NewInstanceStub(klass_reg, dst, klass, info, C1StubId::new_instance_id);
     __ branch(lir_cond_always, slow_path);
     __ branch_destination(slow_path->continuation());
   }
@@ -1479,7 +1479,7 @@ void LIRGenerator::do_RegisterFinalizer(Intrinsic* x) {
   args->append(receiver.result());
   CodeEmitInfo* info = state_for(x, x->state());
   call_runtime(&signature, args,
-               CAST_FROM_FN_PTR(address, Runtime1::entry_for(Runtime1::register_finalizer_id)),
+               CAST_FROM_FN_PTR(address, Runtime1::entry_for(C1StubId::register_finalizer_id)),
                voidType, info);
 
   set_no_result(x);
diff --git a/src/hotspot/share/c1/c1_Runtime1.cpp b/src/hotspot/share/c1/c1_Runtime1.cpp
index 8524f37177b45..5b44d5c0f1983 100644
--- a/src/hotspot/share/c1/c1_Runtime1.cpp
+++ b/src/hotspot/share/c1/c1_Runtime1.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999, 2023, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 1999, 2024, Oracle and/or its affiliates. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -109,10 +109,13 @@ void StubAssembler::set_num_rt_args(int args) {
 
 // Implementation of Runtime1
 
-CodeBlob* Runtime1::_blobs[Runtime1::number_of_ids];
+CodeBlob* Runtime1::_blobs[(int)C1StubId::NUM_STUBIDS];
+
+#define C1_BLOB_NAME_DEFINE(name)  "C1 Runtime " # name "_blob",
 const char *Runtime1::_blob_names[] = {
-  RUNTIME1_STUBS(STUB_NAME, LAST_STUB_NAME)
+  C1_STUBS_DO(C1_BLOB_NAME_DEFINE)
 };
+#undef C1_STUB_NAME_DEFINE
 
 #ifndef PRODUCT
 // statistics
@@ -190,17 +193,17 @@ static void deopt_caller(JavaThread* current) {
   }
 }
 
-class StubIDStubAssemblerCodeGenClosure: public StubAssemblerCodeGenClosure {
+class C1StubIdStubAssemblerCodeGenClosure: public StubAssemblerCodeGenClosure {
  private:
-  Runtime1::StubID _id;
+  C1StubId _id;
  public:
-  StubIDStubAssemblerCodeGenClosure(Runtime1::StubID id) : _id(id) {}
+  C1StubIdStubAssemblerCodeGenClosure(C1StubId id) : _id(id) {}
   virtual OopMapSet* generate_code(StubAssembler* sasm) {
     return Runtime1::generate_code_for(_id, sasm);
   }
 };
 
-CodeBlob* Runtime1::generate_blob(BufferBlob* buffer_blob, int stub_id, const char* name, bool expect_oop_map, StubAssemblerCodeGenClosure* cl) {
+CodeBlob* Runtime1::generate_blob(BufferBlob* buffer_blob, C1StubId id, const char* name, bool expect_oop_map, StubAssemblerCodeGenClosure* cl) {
   ResourceMark rm;
   // create code buffer for code storage
   CodeBuffer code(buffer_blob);
@@ -212,7 +215,7 @@ CodeBlob* Runtime1::generate_blob(BufferBlob* buffer_blob, int stub_id, const ch
   Compilation::setup_code_buffer(&code, 0);
 
   // create assembler for code generation
-  StubAssembler* sasm = new StubAssembler(&code, name, stub_id);
+  StubAssembler* sasm = new StubAssembler(&code, name, (int)id);
   // generate code for runtime stub
   oop_maps = cl->generate_code(sasm);
   assert(oop_maps == nullptr || sasm->frame_size() != no_frame_size,
@@ -237,40 +240,41 @@ CodeBlob* Runtime1::generate_blob(BufferBlob* buffer_blob, int stub_id, const ch
   return blob;
 }
 
-void Runtime1::generate_blob_for(BufferBlob* buffer_blob, StubID id) {
-  assert(0 <= id && id < number_of_ids, "illegal stub id");
+void Runtime1::generate_blob_for(BufferBlob* buffer_blob, C1StubId id) {
+  assert(C1StubId::NO_STUBID < id && id < C1StubId::NUM_STUBIDS, "illegal stub id");
   bool expect_oop_map = true;
 #ifdef ASSERT
   // Make sure that stubs that need oopmaps have them
   switch (id) {
     // These stubs don't need to have an oopmap
-  case dtrace_object_alloc_id:
-  case slow_subtype_check_id:
-  case fpu2long_stub_id:
-  case unwind_exception_id:
-  case counter_overflow_id:
+  case C1StubId::dtrace_object_alloc_id:
+  case C1StubId::slow_subtype_check_id:
+  case C1StubId::fpu2long_stub_id:
+  case C1StubId::unwind_exception_id:
+  case C1StubId::counter_overflow_id:
     expect_oop_map = false;
     break;
   default:
     break;
   }
 #endif
-  StubIDStubAssemblerCodeGenClosure cl(id);
+  C1StubIdStubAssemblerCodeGenClosure cl(id);
   CodeBlob* blob = generate_blob(buffer_blob, id, name_for(id), expect_oop_map, &cl);
   // install blob
-  _blobs[id] = blob;
+  _blobs[(int)id] = blob;
 }
 
 void Runtime1::initialize(BufferBlob* blob) {
   // platform-dependent initialization
   initialize_pd();
   // generate stubs
-  for (int id = 0; id < number_of_ids; id++) generate_blob_for(blob, (StubID)id);
+  int limit = (int)C1StubId::NUM_STUBIDS;
+  for (int id = 0; id < limit; id++) generate_blob_for(blob, (C1StubId)id);
   // printing
 #ifndef PRODUCT
   if (PrintSimpleStubs) {
     ResourceMark rm;
-    for (int id = 0; id < number_of_ids; id++) {
+    for (int id = 0; id < limit; id++) {
       _blobs[id]->print();
       if (_blobs[id]->oop_maps() != nullptr) {
         _blobs[id]->oop_maps()->print();
@@ -282,20 +286,22 @@ void Runtime1::initialize(BufferBlob* blob) {
   bs->generate_c1_runtime_stubs(blob);
 }
 
-CodeBlob* Runtime1::blob_for(StubID id) {
-  assert(0 <= id && id < number_of_ids, "illegal stub id");
-  return _blobs[id];
+CodeBlob* Runtime1::blob_for(C1StubId id) {
+  assert(C1StubId::NO_STUBID < id && id < C1StubId::NUM_STUBIDS, "illegal stub id");
+  return _blobs[(int)id];
 }
 
 
-const char* Runtime1::name_for(StubID id) {
-  assert(0 <= id && id < number_of_ids, "illegal stub id");
-  return _blob_names[id];
+const char* Runtime1::name_for(C1StubId id) {
+  assert(C1StubId::NO_STUBID < id && id < C1StubId::NUM_STUBIDS, "illegal stub id");
+  return _blob_names[(int)id];
 }
 
 const char* Runtime1::name_for_address(address entry) {
-  for (int id = 0; id < number_of_ids; id++) {
-    if (entry == entry_for((StubID)id)) return name_for((StubID)id);
+  int limit = (int)C1StubId::NUM_STUBIDS;
+  for (int i = 0; i < limit; i++) {
+    C1StubId id = (C1StubId)i;
+    if (entry == entry_for(id)) return name_for(id);
   }
 
 #define FUNCTION_CASE(a, f) \
@@ -425,8 +431,8 @@ JRT_ENTRY(void, Runtime1::new_multi_array(JavaThread* current, Klass* klass, int
 JRT_END
 
 
-JRT_ENTRY(void, Runtime1::unimplemented_entry(JavaThread* current, StubID id))
-  tty->print_cr("Runtime1::entry_for(%d) returned unimplemented entry point", id);
+JRT_ENTRY(void, Runtime1::unimplemented_entry(JavaThread* current, C1StubId id))
+  tty->print_cr("Runtime1::entry_for(%d) returned unimplemented entry point", (int)id);
 JRT_END
 
 
@@ -525,8 +531,8 @@ JRT_ENTRY_NO_ASYNC(static address, exception_handler_for_pc_helper(JavaThread* c
   // This function is called when we are about to throw an exception. Therefore,
   // we have to poll the stack watermark barrier to make sure that not yet safe
   // stack frames are made safe before returning into them.
-  if (current->last_frame().cb() == Runtime1::blob_for(Runtime1::handle_exception_from_callee_id)) {
-    // The Runtime1::handle_exception_from_callee_id handler is invoked after the
+  if (current->last_frame().cb() == Runtime1::blob_for(C1StubId::handle_exception_from_callee_id)) {
+    // The C1StubId::handle_exception_from_callee_id handler is invoked after the
     // frame has been unwound. It instead builds its own stub frame, to call the
     // runtime. But the throwing frame has already been unwound here.
     StackWatermarkSet::after_unwind(current);
@@ -922,7 +928,7 @@ static Klass* resolve_field_return_klass(const methodHandle& caller, int bci, TR
 // Therefore, if there is any chance of a race condition, we try to
 // patch only naturally aligned words, as single, full-word writes.
 
-JRT_ENTRY(void, Runtime1::patch_code(JavaThread* current, Runtime1::StubID stub_id ))
+JRT_ENTRY(void, Runtime1::patch_code(JavaThread* current, C1StubId stub_id ))
 #ifndef PRODUCT
   if (PrintC1Statistics) {
     _patch_code_slowcase_cnt++;
@@ -959,9 +965,9 @@ JRT_ENTRY(void, Runtime1::patch_code(JavaThread* current, Runtime1::StubID stub_
   Handle mirror(current, nullptr); // oop needed by load_mirror_patching code
   Handle appendix(current, nullptr); // oop needed by appendix_patching code
   bool load_klass_or_mirror_patch_id =
-    (stub_id == Runtime1::load_klass_patching_id || stub_id == Runtime1::load_mirror_patching_id);
+    (stub_id == C1StubId::load_klass_patching_id || stub_id == C1StubId::load_mirror_patching_id);
 
-  if (stub_id == Runtime1::access_field_patching_id) {
+  if (stub_id == C1StubId::access_field_patching_id) {
 
     Bytecode_field field_access(caller_method, bci);
     fieldDescriptor result; // initialize class if needed
@@ -1044,7 +1050,7 @@ JRT_ENTRY(void, Runtime1::patch_code(JavaThread* current, Runtime1::StubID stub_
       default: fatal("unexpected bytecode for load_klass_or_mirror_patch_id");
     }
     load_klass = k;
-  } else if (stub_id == load_appendix_patching_id) {
+  } else if (stub_id == C1StubId::load_appendix_patching_id) {
     Bytecode_invoke bytecode(caller_method, bci);
     Bytecodes::Code bc = bytecode.invoke_code();
 
@@ -1128,7 +1134,7 @@ JRT_ENTRY(void, Runtime1::patch_code(JavaThread* current, Runtime1::StubID stub_
         if (TracePatching) {
           ttyLocker ttyl;
           tty->print_cr(" Patching %s at bci %d at address " INTPTR_FORMAT "  (%s)", Bytecodes::name(code), bci,
-                        p2i(instr_pc), (stub_id == Runtime1::access_field_patching_id) ? "field" : "klass");
+                        p2i(instr_pc), (stub_id == C1StubId::access_field_patching_id) ? "field" : "klass");
           nmethod* caller_code = CodeCache::find_nmethod(caller_frame.pc());
           assert(caller_code != nullptr, "nmethod not found");
 
@@ -1144,7 +1150,7 @@ JRT_ENTRY(void, Runtime1::patch_code(JavaThread* current, Runtime1::StubID stub_
         }
         // depending on the code below, do_patch says whether to copy the patch body back into the nmethod
         bool do_patch = true;
-        if (stub_id == Runtime1::access_field_patching_id) {
+        if (stub_id == C1StubId::access_field_patching_id) {
           // The offset may not be correct if the class was not loaded at code generation time.
           // Set it now.
           NativeMovRegMem* n_move = nativeMovRegMem_at(copy_buff);
@@ -1170,7 +1176,7 @@ JRT_ENTRY(void, Runtime1::patch_code(JavaThread* current, Runtime1::StubID stub_
             assert(n_copy->data() == 0 ||
                    n_copy->data() == (intptr_t)Universe::non_oop_word(),
                    "illegal init value");
-            if (stub_id == Runtime1::load_klass_patching_id) {
+            if (stub_id == C1StubId::load_klass_patching_id) {
               assert(load_klass != nullptr, "klass not set");
               n_copy->set_data((intx) (load_klass));
             } else {
@@ -1182,7 +1188,7 @@ JRT_ENTRY(void, Runtime1::patch_code(JavaThread* current, Runtime1::StubID stub_
               Disassembler::decode(copy_buff, copy_buff + *byte_count, tty);
             }
           }
-        } else if (stub_id == Runtime1::load_appendix_patching_id) {
+        } else if (stub_id == C1StubId::load_appendix_patching_id) {
           NativeMovConstReg* n_copy = nativeMovConstReg_at(copy_buff);
           assert(n_copy->data() == 0 ||
                  n_copy->data() == (intptr_t)Universe::non_oop_word(),
@@ -1201,7 +1207,7 @@ JRT_ENTRY(void, Runtime1::patch_code(JavaThread* current, Runtime1::StubID stub_
           // first replace the tail, then the call
 #ifdef ARM
           if((load_klass_or_mirror_patch_id ||
-              stub_id == Runtime1::load_appendix_patching_id) &&
+              stub_id == C1StubId::load_appendix_patching_id) &&
               nativeMovConstReg_at(copy_buff)->is_pc_relative()) {
             nmethod* nm = CodeCache::find_nmethod(instr_pc);
             address addr = nullptr;
@@ -1209,13 +1215,13 @@ JRT_ENTRY(void, Runtime1::patch_code(JavaThread* current, Runtime1::StubID stub_
             RelocIterator mds(nm, copy_buff, copy_buff + 1);
             while (mds.next()) {
               if (mds.type() == relocInfo::oop_type) {
-                assert(stub_id == Runtime1::load_mirror_patching_id ||
-                       stub_id == Runtime1::load_appendix_patching_id, "wrong stub id");
+                assert(stub_id == C1StubId::load_mirror_patching_id ||
+                       stub_id == C1StubId::load_appendix_patching_id, "wrong stub id");
                 oop_Relocation* r = mds.oop_reloc();
                 addr = (address)r->oop_addr();
                 break;
               } else if (mds.type() == relocInfo::metadata_type) {
-                assert(stub_id == Runtime1::load_klass_patching_id, "wrong stub id");
+                assert(stub_id == C1StubId::load_klass_patching_id, "wrong stub id");
                 metadata_Relocation* r = mds.metadata_reloc();
                 addr = (address)r->metadata_addr();
                 break;
@@ -1238,9 +1244,9 @@ JRT_ENTRY(void, Runtime1::patch_code(JavaThread* current, Runtime1::StubID stub_
           NativeGeneralJump::replace_mt_safe(instr_pc, copy_buff);
 
           if (load_klass_or_mirror_patch_id ||
-              stub_id == Runtime1::load_appendix_patching_id) {
+              stub_id == C1StubId::load_appendix_patching_id) {
             relocInfo::relocType rtype =
-              (stub_id == Runtime1::load_klass_patching_id) ?
+              (stub_id == C1StubId::load_klass_patching_id) ?
                                    relocInfo::metadata_type :
                                    relocInfo::oop_type;
             // update relocInfo to metadata
@@ -1278,9 +1284,9 @@ JRT_END
 
 #else // DEOPTIMIZE_WHEN_PATCHING
 
-static bool is_patching_needed(JavaThread* current, Runtime1::StubID stub_id) {
-  if (stub_id == Runtime1::load_klass_patching_id ||
-      stub_id == Runtime1::load_mirror_patching_id) {
+static bool is_patching_needed(JavaThread* current, C1StubId stub_id) {
+  if (stub_id == C1StubId::load_klass_patching_id ||
+      stub_id == C1StubId::load_mirror_patching_id) {
     // last java frame on stack
     vframeStream vfst(current, true);
     assert(!vfst.at_end(), "Java frame must exist");
@@ -1309,7 +1315,7 @@ static bool is_patching_needed(JavaThread* current, Runtime1::StubID stub_id) {
   return true;
 }
 
-void Runtime1::patch_code(JavaThread* current, Runtime1::StubID stub_id) {
+void Runtime1::patch_code(JavaThread* current, C1StubId stub_id) {
 #ifndef PRODUCT
   if (PrintC1Statistics) {
     _patch_code_slowcase_cnt++;
@@ -1364,7 +1370,7 @@ int Runtime1::move_klass_patching(JavaThread* current) {
   {
     // Enter VM mode
     ResetNoHandleMark rnhm;
-    patch_code(current, load_klass_patching_id);
+    patch_code(current, C1StubId::load_klass_patching_id);
   }
   // Back in JAVA, use no oops DON'T safepoint
 
@@ -1381,7 +1387,7 @@ int Runtime1::move_mirror_patching(JavaThread* current) {
   {
     // Enter VM mode
     ResetNoHandleMark rnhm;
-    patch_code(current, load_mirror_patching_id);
+    patch_code(current, C1StubId::load_mirror_patching_id);
   }
   // Back in JAVA, use no oops DON'T safepoint
 
@@ -1398,7 +1404,7 @@ int Runtime1::move_appendix_patching(JavaThread* current) {
   {
     // Enter VM mode
     ResetNoHandleMark rnhm;
-    patch_code(current, load_appendix_patching_id);
+    patch_code(current, C1StubId::load_appendix_patching_id);
   }
   // Back in JAVA, use no oops DON'T safepoint
 
@@ -1425,7 +1431,7 @@ int Runtime1::access_field_patching(JavaThread* current) {
   {
     // Enter VM mode
     ResetNoHandleMark rnhm;
-    patch_code(current, access_field_patching_id);
+    patch_code(current, C1StubId::access_field_patching_id);
   }
   // Back in JAVA, use no oops DON'T safepoint
 
diff --git a/src/hotspot/share/c1/c1_Runtime1.hpp b/src/hotspot/share/c1/c1_Runtime1.hpp
index 2e4c9f8a73379..330c40675041b 100644
--- a/src/hotspot/share/c1/c1_Runtime1.hpp
+++ b/src/hotspot/share/c1/c1_Runtime1.hpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999, 2023, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 1999, 2024, Oracle and/or its affiliates. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -30,6 +30,7 @@
 #include "interpreter/interpreter.hpp"
 #include "memory/allStatic.hpp"
 #include "runtime/deoptimization.hpp"
+#include "runtime/stubDeclarations.hpp"
 
 class StubAssembler;
 
@@ -37,61 +38,26 @@ class StubAssembler;
 // runtime routines needed by code code generated
 // by the Compiler1.
 
-#define RUNTIME1_STUBS(stub, last_entry) \
-  stub(dtrace_object_alloc)          \
-  stub(unwind_exception)             \
-  stub(forward_exception)            \
-  stub(throw_range_check_failed)       /* throws ArrayIndexOutOfBoundsException */ \
-  stub(throw_index_exception)          /* throws IndexOutOfBoundsException */ \
-  stub(throw_div0_exception)         \
-  stub(throw_null_pointer_exception) \
-  stub(register_finalizer)           \
-  stub(new_instance)                 \
-  stub(fast_new_instance)            \
-  stub(fast_new_instance_init_check) \
-  stub(new_type_array)               \
-  stub(new_object_array)             \
-  stub(new_multi_array)              \
-  stub(handle_exception_nofpu)         /* optimized version that does not preserve fpu registers */ \
-  stub(handle_exception)             \
-  stub(handle_exception_from_callee) \
-  stub(throw_array_store_exception)  \
-  stub(throw_class_cast_exception)   \
-  stub(throw_incompatible_class_change_error)   \
-  stub(slow_subtype_check)           \
-  stub(monitorenter)                 \
-  stub(monitorenter_nofpu)             /* optimized version that does not preserve fpu registers */ \
-  stub(monitorexit)                  \
-  stub(monitorexit_nofpu)              /* optimized version that does not preserve fpu registers */ \
-  stub(deoptimize)                   \
-  stub(access_field_patching)        \
-  stub(load_klass_patching)          \
-  stub(load_mirror_patching)         \
-  stub(load_appendix_patching)       \
-  stub(fpu2long_stub)                \
-  stub(counter_overflow)             \
-  stub(predicate_failed_trap)        \
-  last_entry(number_of_ids)
-
-#define DECLARE_STUB_ID(x)       x ## _id ,
-#define DECLARE_LAST_STUB_ID(x)  x
-#define STUB_NAME(x)             #x " Runtime1 stub",
-#define LAST_STUB_NAME(x)        #x " Runtime1 stub"
-
 class StubAssemblerCodeGenClosure: public Closure {
  public:
   virtual OopMapSet* generate_code(StubAssembler* sasm) = 0;
 };
 
+// define C1StubId enum tags: unwind_exception_id etc
+
+#define C1_STUB_ID_ENUM_DECLARE(name) STUB_ID_NAME(name),
+enum class C1StubId :int {
+  NO_STUBID = -1,
+  C1_STUBS_DO(C1_STUB_ID_ENUM_DECLARE)
+  NUM_STUBIDS
+};
+#undef C1_STUB_ID_ENUM_DECLARE
+
 class Runtime1: public AllStatic {
   friend class VMStructs;
   friend class ArrayCopyStub;
 
- public:
-  enum StubID {
-    RUNTIME1_STUBS(DECLARE_STUB_ID, DECLARE_LAST_STUB_ID)
-  };
-
+public:
   // statistics
 #ifndef PRODUCT
   static uint _generic_arraycopystub_cnt;
@@ -115,17 +81,17 @@ class Runtime1: public AllStatic {
 #endif
 
  private:
-  static CodeBlob* _blobs[number_of_ids];
+  static CodeBlob* _blobs[(int)C1StubId::NUM_STUBIDS];
   static const char* _blob_names[];
 
   // stub generation
  public:
-  static CodeBlob*  generate_blob(BufferBlob* buffer_blob, int stub_id, const char* name, bool expect_oop_map, StubAssemblerCodeGenClosure *cl);
-  static void       generate_blob_for(BufferBlob* blob, StubID id);
-  static OopMapSet* generate_code_for(StubID id, StubAssembler* sasm);
+  static CodeBlob*  generate_blob(BufferBlob* buffer_blob, C1StubId id, const char* name, bool expect_oop_map, StubAssemblerCodeGenClosure *cl);
+  static void       generate_blob_for(BufferBlob* blob, C1StubId id);
+  static OopMapSet* generate_code_for(C1StubId id, StubAssembler* sasm);
  private:
   static OopMapSet* generate_exception_throw(StubAssembler* sasm, address target, bool has_argument);
-  static OopMapSet* generate_handle_exception(StubID id, StubAssembler* sasm);
+  static OopMapSet* generate_handle_exception(C1StubId id, StubAssembler* sasm);
   static void       generate_unwind_exception(StubAssembler *sasm);
   static OopMapSet* generate_patching(StubAssembler* sasm, address target);
 
@@ -140,7 +106,7 @@ class Runtime1: public AllStatic {
 
   static address counter_overflow(JavaThread* current, int bci, Method* method);
 
-  static void unimplemented_entry(JavaThread* current, StubID id);
+  static void unimplemented_entry(JavaThread* current, C1StubId id);
 
   static address exception_handler_for_pc(JavaThread* current);
 
@@ -162,7 +128,7 @@ class Runtime1: public AllStatic {
   static int move_mirror_patching(JavaThread* current);
   static int move_appendix_patching(JavaThread* current);
 
-  static void patch_code(JavaThread* current, StubID stub_id);
+  static void patch_code(JavaThread* current, C1StubId stub_id);
 
  public:
   // initialization
@@ -170,9 +136,9 @@ class Runtime1: public AllStatic {
   static void initialize_pd();
 
   // stubs
-  static CodeBlob* blob_for (StubID id);
-  static address   entry_for(StubID id)          { return blob_for(id)->code_begin(); }
-  static const char* name_for (StubID id);
+  static CodeBlob* blob_for (C1StubId id);
+  static address   entry_for(C1StubId id)          { return blob_for(id)->code_begin(); }
+  static const char* name_for (C1StubId id);
   static const char* name_for_address(address entry);
 
   // platform might add runtime names.
diff --git a/src/hotspot/share/gc/g1/c1/g1BarrierSetC1.cpp b/src/hotspot/share/gc/g1/c1/g1BarrierSetC1.cpp
index f82b5cfcc556a..449ff2e4acf8b 100644
--- a/src/hotspot/share/gc/g1/c1/g1BarrierSetC1.cpp
+++ b/src/hotspot/share/gc/g1/c1/g1BarrierSetC1.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018, 2023, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2018, 2024, Oracle and/or its affiliates. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -219,8 +219,8 @@ class C1G1PostBarrierCodeGenClosure : public StubAssemblerCodeGenClosure {
 void G1BarrierSetC1::generate_c1_runtime_stubs(BufferBlob* buffer_blob) {
   C1G1PreBarrierCodeGenClosure pre_code_gen_cl;
   C1G1PostBarrierCodeGenClosure post_code_gen_cl;
-  _pre_barrier_c1_runtime_code_blob = Runtime1::generate_blob(buffer_blob, -1, "g1_pre_barrier_slow",
+  _pre_barrier_c1_runtime_code_blob = Runtime1::generate_blob(buffer_blob, C1StubId::NO_STUBID, "g1_pre_barrier_slow",
                                                               false, &pre_code_gen_cl);
-  _post_barrier_c1_runtime_code_blob = Runtime1::generate_blob(buffer_blob, -1, "g1_post_barrier_slow",
+  _post_barrier_c1_runtime_code_blob = Runtime1::generate_blob(buffer_blob, C1StubId::NO_STUBID, "g1_post_barrier_slow",
                                                                false, &post_code_gen_cl);
 }
diff --git a/src/hotspot/share/gc/shenandoah/c1/shenandoahBarrierSetC1.cpp b/src/hotspot/share/gc/shenandoah/c1/shenandoahBarrierSetC1.cpp
index 6b104e7e19935..0e8b02d247e9b 100644
--- a/src/hotspot/share/gc/shenandoah/c1/shenandoahBarrierSetC1.cpp
+++ b/src/hotspot/share/gc/shenandoah/c1/shenandoahBarrierSetC1.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018, 2022, Red Hat, Inc. All rights reserved.
+ * Copyright (c) 2018, 2024, Red Hat, Inc. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -263,27 +263,27 @@ class C1ShenandoahLoadReferenceBarrierCodeGenClosure : public StubAssemblerCodeG
 
 void ShenandoahBarrierSetC1::generate_c1_runtime_stubs(BufferBlob* buffer_blob) {
   C1ShenandoahPreBarrierCodeGenClosure pre_code_gen_cl;
-  _pre_barrier_c1_runtime_code_blob = Runtime1::generate_blob(buffer_blob, -1,
+  _pre_barrier_c1_runtime_code_blob = Runtime1::generate_blob(buffer_blob, C1StubId::NO_STUBID,
                                                               "shenandoah_pre_barrier_slow",
                                                               false, &pre_code_gen_cl);
   if (ShenandoahLoadRefBarrier) {
     C1ShenandoahLoadReferenceBarrierCodeGenClosure lrb_strong_code_gen_cl(ON_STRONG_OOP_REF);
-    _load_reference_barrier_strong_rt_code_blob = Runtime1::generate_blob(buffer_blob, -1,
+    _load_reference_barrier_strong_rt_code_blob = Runtime1::generate_blob(buffer_blob, C1StubId::NO_STUBID,
                                                                   "shenandoah_load_reference_barrier_strong_slow",
                                                                   false, &lrb_strong_code_gen_cl);
 
     C1ShenandoahLoadReferenceBarrierCodeGenClosure lrb_strong_native_code_gen_cl(ON_STRONG_OOP_REF | IN_NATIVE);
-    _load_reference_barrier_strong_native_rt_code_blob = Runtime1::generate_blob(buffer_blob, -1,
+    _load_reference_barrier_strong_native_rt_code_blob = Runtime1::generate_blob(buffer_blob, C1StubId::NO_STUBID,
                                                                           "shenandoah_load_reference_barrier_strong_native_slow",
                                                                           false, &lrb_strong_native_code_gen_cl);
 
     C1ShenandoahLoadReferenceBarrierCodeGenClosure lrb_weak_code_gen_cl(ON_WEAK_OOP_REF);
-    _load_reference_barrier_weak_rt_code_blob = Runtime1::generate_blob(buffer_blob, -1,
+    _load_reference_barrier_weak_rt_code_blob = Runtime1::generate_blob(buffer_blob, C1StubId::NO_STUBID,
                                                                           "shenandoah_load_reference_barrier_weak_slow",
                                                                           false, &lrb_weak_code_gen_cl);
 
     C1ShenandoahLoadReferenceBarrierCodeGenClosure lrb_phantom_code_gen_cl(ON_PHANTOM_OOP_REF | IN_NATIVE);
-    _load_reference_barrier_phantom_rt_code_blob = Runtime1::generate_blob(buffer_blob, -1,
+    _load_reference_barrier_phantom_rt_code_blob = Runtime1::generate_blob(buffer_blob, C1StubId::NO_STUBID,
                                                                            "shenandoah_load_reference_barrier_phantom_slow",
                                                                            false, &lrb_phantom_code_gen_cl);
   }
diff --git a/src/hotspot/share/gc/x/c1/xBarrierSetC1.cpp b/src/hotspot/share/gc/x/c1/xBarrierSetC1.cpp
index caf7eb2e514e9..6f64392cefced 100644
--- a/src/hotspot/share/gc/x/c1/xBarrierSetC1.cpp
+++ b/src/hotspot/share/gc/x/c1/xBarrierSetC1.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2015, 2022, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2015, 2024, Oracle and/or its affiliates. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -225,7 +225,7 @@ class XLoadBarrierRuntimeStubCodeGenClosure : public StubAssemblerCodeGenClosure
 
 static address generate_c1_runtime_stub(BufferBlob* blob, DecoratorSet decorators, const char* name) {
   XLoadBarrierRuntimeStubCodeGenClosure cl(decorators);
-  CodeBlob* const code_blob = Runtime1::generate_blob(blob, -1 /* stub_id */, name, false /* expect_oop_map*/, &cl);
+  CodeBlob* const code_blob = Runtime1::generate_blob(blob, C1StubId::NO_STUBID /* stub_id */, name, false /* expect_oop_map*/, &cl);
   return code_blob->code_begin();
 }
 
diff --git a/src/hotspot/share/gc/z/c1/zBarrierSetC1.cpp b/src/hotspot/share/gc/z/c1/zBarrierSetC1.cpp
index 68e215e9ebca0..9c16714b26e3f 100644
--- a/src/hotspot/share/gc/z/c1/zBarrierSetC1.cpp
+++ b/src/hotspot/share/gc/z/c1/zBarrierSetC1.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2015, 2023, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2015, 2024, Oracle and/or its affiliates. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -507,7 +507,7 @@ class ZLoadBarrierRuntimeStubCodeGenClosure : public StubAssemblerCodeGenClosure
 
 static address generate_c1_load_runtime_stub(BufferBlob* blob, DecoratorSet decorators, const char* name) {
   ZLoadBarrierRuntimeStubCodeGenClosure cl(decorators);
-  CodeBlob* const code_blob = Runtime1::generate_blob(blob, -1 /* stub_id */, name, false /* expect_oop_map*/, &cl);
+  CodeBlob* const code_blob = Runtime1::generate_blob(blob, C1StubId::NO_STUBID /* stub_id */, name, false /* expect_oop_map*/, &cl);
   return code_blob->code_begin();
 }
 
@@ -527,7 +527,7 @@ class ZStoreBarrierRuntimeStubCodeGenClosure : public StubAssemblerCodeGenClosur
 
 static address generate_c1_store_runtime_stub(BufferBlob* blob, bool self_healing, const char* name) {
   ZStoreBarrierRuntimeStubCodeGenClosure cl(self_healing);
-  CodeBlob* const code_blob = Runtime1::generate_blob(blob, -1 /* stub_id */, name, false /* expect_oop_map*/, &cl);
+  CodeBlob* const code_blob = Runtime1::generate_blob(blob, C1StubId::NO_STUBID /* stub_id */, name, false /* expect_oop_map*/, &cl);
   return code_blob->code_begin();
 }
 
diff --git a/src/hotspot/share/opto/escape.cpp b/src/hotspot/share/opto/escape.cpp
index eb6887f11baf6..6ab28eaa6eeba 100644
--- a/src/hotspot/share/opto/escape.cpp
+++ b/src/hotspot/share/opto/escape.cpp
@@ -2043,7 +2043,7 @@ void ConnectionGraph::add_call_node(CallNode* call) {
     ciMethod* meth = call->as_CallJava()->method();
     if (meth == nullptr) {
       const char* name = call->as_CallStaticJava()->_name;
-      assert(strncmp(name, "_multianewarray", 15) == 0, "TODO: add failed case check");
+      assert(strncmp(name, "C2 Runtime multianewarray", 25) == 0, "TODO: add failed case check");
       // Returns a newly allocated non-escaped object.
       add_java_object(call, PointsToNode::NoEscape);
       set_not_scalar_replaceable(ptnode_adr(call_idx) NOT_PRODUCT(COMMA "is result of multinewarray"));
@@ -2738,7 +2738,7 @@ int ConnectionGraph::find_init_values_phantom(JavaObjectNode* pta) {
 #ifdef ASSERT
   if (!pta->arraycopy_dst() && alloc->as_CallStaticJava()->method() == nullptr) {
     const char* name = alloc->as_CallStaticJava()->_name;
-    assert(strncmp(name, "_multianewarray", 15) == 0, "sanity");
+    assert(strncmp(name, "C2 Runtime multianewarray", 25) == 0, "sanity");
   }
 #endif
   // Non-escaped allocation returned from Java or runtime call have unknown values in fields.
diff --git a/src/hotspot/share/opto/runtime.cpp b/src/hotspot/share/opto/runtime.cpp
index 54408146d0c26..3dd94f619fd1a 100644
--- a/src/hotspot/share/opto/runtime.cpp
+++ b/src/hotspot/share/opto/runtime.cpp
@@ -88,34 +88,27 @@
 // At command line specify the parameters: -XX:+FullGCALot -XX:FullGCALotStart=100000000
 
 
-
-
-// Compiled code entry points
-address OptoRuntime::_new_instance_Java                           = nullptr;
-address OptoRuntime::_new_array_Java                              = nullptr;
-address OptoRuntime::_new_array_nozero_Java                       = nullptr;
-address OptoRuntime::_multianewarray2_Java                        = nullptr;
-address OptoRuntime::_multianewarray3_Java                        = nullptr;
-address OptoRuntime::_multianewarray4_Java                        = nullptr;
-address OptoRuntime::_multianewarray5_Java                        = nullptr;
-address OptoRuntime::_multianewarrayN_Java                        = nullptr;
-address OptoRuntime::_vtable_must_compile_Java                    = nullptr;
-address OptoRuntime::_complete_monitor_locking_Java               = nullptr;
-address OptoRuntime::_monitor_notify_Java                         = nullptr;
-address OptoRuntime::_monitor_notifyAll_Java                      = nullptr;
-address OptoRuntime::_rethrow_Java                                = nullptr;
-
-address OptoRuntime::_slow_arraycopy_Java                         = nullptr;
-address OptoRuntime::_register_finalizer_Java                     = nullptr;
-#if INCLUDE_JVMTI
-address OptoRuntime::_notify_jvmti_vthread_start                  = nullptr;
-address OptoRuntime::_notify_jvmti_vthread_end                    = nullptr;
-address OptoRuntime::_notify_jvmti_vthread_mount                  = nullptr;
-address OptoRuntime::_notify_jvmti_vthread_unmount                = nullptr;
-#endif
-
-UncommonTrapBlob*   OptoRuntime::_uncommon_trap_blob;
-ExceptionBlob*      OptoRuntime::_exception_blob;
+#define C2_BLOB_FIELD_DEFINE(name, type) \
+  type OptoRuntime:: BLOB_FIELD_NAME(name)  = nullptr;
+#define C2_STUB_FIELD_NAME(name) _ ## name ## _Java
+#define C2_STUB_FIELD_DEFINE(name, f, t, r) \
+  address OptoRuntime:: C2_STUB_FIELD_NAME(name) = nullptr;
+#define C2_JVMTI_STUB_FIELD_DEFINE(name) \
+  address OptoRuntime:: STUB_FIELD_NAME(name) = nullptr;
+C2_STUBS_DO(C2_BLOB_FIELD_DEFINE, C2_STUB_FIELD_DEFINE, C2_JVMTI_STUB_FIELD_DEFINE)
+#undef C2_BLOB_FIELD_DEFINE
+#undef C2_STUB_FIELD_DEFINE
+#undef C2_JVMTI_STUB_FIELD_DEFINE
+
+#define C2_BLOB_NAME_DEFINE(name, type)  "C2 Runtime " # name "_blob",
+#define C2_STUB_NAME_DEFINE(name, f, t, r)  "C2 Runtime " # name,
+#define C2_JVMTI_STUB_NAME_DEFINE(name)  "C2 Runtime " # name,
+const char* OptoRuntime::_stub_names[] = {
+  C2_STUBS_DO(C2_BLOB_NAME_DEFINE, C2_STUB_NAME_DEFINE, C2_JVMTI_STUB_NAME_DEFINE)
+};
+#undef C2_BLOB_NAME_DEFINE
+#undef C2_STUB_NAME_DEFINE
+#undef C2_JVMTI_STUB_NAME_DEFINE
 
 // This should be called in an assertion at the start of OptoRuntime routines
 // which are entered from compiled code (all of them)
@@ -132,46 +125,72 @@ static bool check_compiled_frame(JavaThread* thread) {
 }
 #endif // ASSERT
 
-
+/*
 #define gen(env, var, type_func_gen, c_func, fancy_jump, pass_tls, return_pc) \
   var = generate_stub(env, type_func_gen, CAST_FROM_FN_PTR(address, c_func), #var, fancy_jump, pass_tls, return_pc); \
   if (var == nullptr) { return false; }
+*/
 
-bool OptoRuntime::generate(ciEnv* env) {
+#define GEN_C2_BLOB(name, type)                    \
+  generate_ ## name ## _blob();
+
+// a few helper macros to conjure up generate_stub call arguments
+#define C2_STUB_FIELD_NAME(name) _ ## name ## _Java
+#define C2_STUB_TYPEFUNC(name) name ## _Type
+#define C2_STUB_C_FUNC(name) CAST_FROM_FN_PTR(address, name ## _C)
+#define C2_STUB_NAME(name) stub_name(OptoStubId::name ## _id)
+
+// Almost all the C functions targeted from the generated stubs are
+// implemented locally to OptoRuntime with names that can be generated
+// from the stub name by appending suffix '_C'. However, in two cases
+// a common target method also needs to be called from shared runtime
+// stubs. In these two cases the opto stubs rely on method
+// imlementations defined in class SharedRuntime. The following
+// defines temporarily rebind the generated names to reference the
+// relevant implementations.
+
+#define GEN_C2_STUB(name, fancy_jump, pass_tls, pass_retpc  )         \
+  C2_STUB_FIELD_NAME(name) =                                          \
+    generate_stub(env,                                                  \
+                  C2_STUB_TYPEFUNC(name),                             \
+                  C2_STUB_C_FUNC(name),                               \
+                  C2_STUB_NAME(name),                                 \
+                  fancy_jump,                                           \
+                  pass_tls,                                             \
+                  pass_retpc);                                          \
+  if (C2_STUB_FIELD_NAME(name) == nullptr) { return false; }          \
+
+#define C2_JVMTI_STUB_C_FUNC(name) CAST_FROM_FN_PTR(address, SharedRuntime::name)
+
+#define GEN_C2_JVMTI_STUB(name)                                       \
+  STUB_FIELD_NAME(name) =                                               \
+    generate_stub(env,                                                  \
+                  notify_jvmti_vthread_Type,                            \
+                  C2_JVMTI_STUB_C_FUNC(name),                         \
+                  C2_STUB_NAME(name),                                 \
+                  0,                                                    \
+                  true,                                                 \
+                  false);                                               \
+  if (STUB_FIELD_NAME(name) == nullptr) { return false; }               \
 
-  generate_uncommon_trap_blob();
-  generate_exception_blob();
-
-  // Note: tls: Means fetching the return oop out of the thread-local storage
-  //
-  //   variable/name                       type-function-gen              , runtime method                  ,fncy_jp, tls,retpc
-  // -------------------------------------------------------------------------------------------------------------------------------
-  gen(env, _new_instance_Java              , new_instance_Type            , new_instance_C                  ,    0 , true, false);
-  gen(env, _new_array_Java                 , new_array_Type               , new_array_C                     ,    0 , true, false);
-  gen(env, _new_array_nozero_Java          , new_array_Type               , new_array_nozero_C              ,    0 , true, false);
-  gen(env, _multianewarray2_Java           , multianewarray2_Type         , multianewarray2_C               ,    0 , true, false);
-  gen(env, _multianewarray3_Java           , multianewarray3_Type         , multianewarray3_C               ,    0 , true, false);
-  gen(env, _multianewarray4_Java           , multianewarray4_Type         , multianewarray4_C               ,    0 , true, false);
-  gen(env, _multianewarray5_Java           , multianewarray5_Type         , multianewarray5_C               ,    0 , true, false);
-  gen(env, _multianewarrayN_Java           , multianewarrayN_Type         , multianewarrayN_C               ,    0 , true, false);
-#if INCLUDE_JVMTI
-  gen(env, _notify_jvmti_vthread_start     , notify_jvmti_vthread_Type    , SharedRuntime::notify_jvmti_vthread_start, 0, true, false);
-  gen(env, _notify_jvmti_vthread_end       , notify_jvmti_vthread_Type    , SharedRuntime::notify_jvmti_vthread_end,   0, true, false);
-  gen(env, _notify_jvmti_vthread_mount     , notify_jvmti_vthread_Type    , SharedRuntime::notify_jvmti_vthread_mount, 0, true, false);
-  gen(env, _notify_jvmti_vthread_unmount   , notify_jvmti_vthread_Type    , SharedRuntime::notify_jvmti_vthread_unmount, 0, true, false);
-#endif
-  gen(env, _complete_monitor_locking_Java  , complete_monitor_enter_Type  , SharedRuntime::complete_monitor_locking_C, 0, false, false);
-  gen(env, _monitor_notify_Java            , monitor_notify_Type          , monitor_notify_C                ,    0 , false, false);
-  gen(env, _monitor_notifyAll_Java         , monitor_notify_Type          , monitor_notifyAll_C             ,    0 , false, false);
-  gen(env, _rethrow_Java                   , rethrow_Type                 , rethrow_C                       ,    2 , true , true );
+bool OptoRuntime::generate(ciEnv* env) {
 
-  gen(env, _slow_arraycopy_Java            , slow_arraycopy_Type          , SharedRuntime::slow_arraycopy_C ,    0 , false, false);
-  gen(env, _register_finalizer_Java        , register_finalizer_Type      , register_finalizer              ,    0 , false, false);
+  C2_STUBS_DO(GEN_C2_BLOB, GEN_C2_STUB, GEN_C2_JVMTI_STUB)
 
   return true;
 }
 
-#undef gen
+#undef GEN_C2_BLOB
+
+#undef C2_STUB_FIELD_NAME
+#undef C2_STUB_TYPEFUNC
+#undef C2_STUB_C_FUNC
+#undef C2_STUB_NAME
+#undef GEN_C2_STUB
+
+#undef C2_JVMTI_STUB_C_FUNC
+#undef GEN_C2_JVMTI_STUB
+// #undef gen
 
 
 // Helper method to do generation of RunTimeStub's
@@ -201,6 +220,19 @@ const char* OptoRuntime::stub_name(address entry) {
 #endif
 }
 
+// local methods passed as arguments to stub generator that forward
+// control to corresponding JRT methods of SharedRuntime
+
+void OptoRuntime::slow_arraycopy_C(oopDesc* src,  jint src_pos,
+                                   oopDesc* dest, jint dest_pos,
+                                   jint length, JavaThread* thread) {
+  SharedRuntime::slow_arraycopy_C(src,  src_pos, dest, dest_pos, length, thread);
+}
+
+void OptoRuntime::complete_monitor_locking_C(oopDesc* obj, BasicLock* lock, JavaThread* current) {
+  SharedRuntime::complete_monitor_locking_C(obj, lock, current);
+}
+
 
 //=============================================================================
 // Opto compiler runtime routines
@@ -529,6 +561,10 @@ const TypeFunc *OptoRuntime::new_array_Type() {
   return TypeFunc::make(domain, range);
 }
 
+const TypeFunc *OptoRuntime::new_array_nozero_Type() {
+  return new_array_Type();
+}
+
 const TypeFunc *OptoRuntime::multianewarray_Type(int ndim) {
   // create input type (domain)
   const int nargs = ndim + 1;
@@ -607,6 +643,9 @@ const TypeFunc *OptoRuntime::complete_monitor_enter_Type() {
   return TypeFunc::make(domain,range);
 }
 
+const TypeFunc *OptoRuntime::complete_monitor_locking_Type() {
+  return complete_monitor_enter_Type();
+}
 
 //-----------------------------------------------------------------------------
 const TypeFunc *OptoRuntime::complete_monitor_exit_Type() {
@@ -637,6 +676,10 @@ const TypeFunc *OptoRuntime::monitor_notify_Type() {
   return TypeFunc::make(domain, range);
 }
 
+const TypeFunc *OptoRuntime::monitor_notifyAll_Type() {
+  return monitor_notify_Type();
+}
+
 const TypeFunc* OptoRuntime::flush_windows_Type() {
   // create input type (domain)
   const Type** fields = TypeTuple::fields(1);
@@ -1827,7 +1870,7 @@ const TypeFunc *OptoRuntime::dtrace_object_alloc_Type() {
 }
 
 
-JRT_ENTRY_NO_ASYNC(void, OptoRuntime::register_finalizer(oopDesc* obj, JavaThread* current))
+JRT_ENTRY_NO_ASYNC(void, OptoRuntime::register_finalizer_C(oopDesc* obj, JavaThread* current))
   assert(oopDesc::is_oop(obj), "must be a valid oop");
   assert(obj->klass()->has_finalizer(), "shouldn't be here otherwise");
   InstanceKlass::register_finalizer(instanceOop(obj), CHECK);
diff --git a/src/hotspot/share/opto/runtime.hpp b/src/hotspot/share/opto/runtime.hpp
index 34c2780a2f8d7..da08fc0f38660 100644
--- a/src/hotspot/share/opto/runtime.hpp
+++ b/src/hotspot/share/opto/runtime.hpp
@@ -30,6 +30,7 @@
 #include "opto/optoreg.hpp"
 #include "opto/type.hpp"
 #include "runtime/deoptimization.hpp"
+#include "runtime/stubDeclarations.hpp"
 #include "runtime/vframe.hpp"
 
 //------------------------------OptoRuntime------------------------------------
@@ -98,37 +99,46 @@ class NamedCounter : public CHeapObj<mtCompiler> {
 
 typedef const TypeFunc*(*TypeFunc_generator)();
 
+// define OptoStubId enum tags: uncommon_trap_id etc
+
+#define C2_BLOB_ID_ENUM_DECLARE(name, type) STUB_ID_NAME(name),
+#define C2_STUB_ID_ENUM_DECLARE(name, f, t, r) STUB_ID_NAME(name),
+#define C2_JVMTI_STUB_ID_ENUM_DECLARE(name) STUB_ID_NAME(name),
+enum class OptoStubId :int {
+  NO_STUBID = -1,
+  C2_STUBS_DO(C2_BLOB_ID_ENUM_DECLARE, C2_STUB_ID_ENUM_DECLARE, C2_JVMTI_STUB_ID_ENUM_DECLARE)
+  NUM_STUBIDS
+};
+#undef C2_BLOB_ID_ENUM_DECLARE
+#undef C2_STUB_ID_ENUM_DECLARE
+#undef C2_JVMTI_STUB_ID_ENUM_DECLARE
+
 class OptoRuntime : public AllStatic {
   friend class Matcher;  // allow access to stub names
 
  private:
+  // declare opto stub address/blob holder static fields
+#define C2_BLOB_FIELD_DECLARE(name, type) \
+  static type        BLOB_FIELD_NAME(name);
+#define C2_STUB_FIELD_NAME(name) _ ## name ## _Java
+#define C2_STUB_FIELD_DECLARE(name, f, t, r) \
+  static address     C2_STUB_FIELD_NAME(name) ;
+#define C2_JVMTI_STUB_FIELD_DECLARE(name) \
+  static address     STUB_FIELD_NAME(name);
+
+  C2_STUBS_DO(C2_BLOB_FIELD_DECLARE, C2_STUB_FIELD_DECLARE, C2_JVMTI_STUB_FIELD_DECLARE)
+
+#undef C2_BLOB_FIELD_DECLARE
+#undef C2_STUB_FIELD_NAME
+#undef C2_STUB_FIELD_DECLARE
+#undef C2_JVMTI_STUB_FIELD_DECLARE
+
+  // Stub names indexed by sharedStubId
+  static const char *_stub_names[];
+
   // define stubs
   static address generate_stub(ciEnv* ci_env, TypeFunc_generator gen, address C_function, const char* name, int is_fancy_jump, bool pass_tls, bool return_pc);
 
-  // References to generated stubs
-  static address _new_instance_Java;
-  static address _new_array_Java;
-  static address _new_array_nozero_Java;
-  static address _multianewarray2_Java;
-  static address _multianewarray3_Java;
-  static address _multianewarray4_Java;
-  static address _multianewarray5_Java;
-  static address _multianewarrayN_Java;
-  static address _vtable_must_compile_Java;
-  static address _complete_monitor_locking_Java;
-  static address _rethrow_Java;
-  static address _monitor_notify_Java;
-  static address _monitor_notifyAll_Java;
-
-  static address _slow_arraycopy_Java;
-  static address _register_finalizer_Java;
-#if INCLUDE_JVMTI
-  static address _notify_jvmti_vthread_start;
-  static address _notify_jvmti_vthread_end;
-  static address _notify_jvmti_vthread_mount;
-  static address _notify_jvmti_vthread_unmount;
-#endif
-
   //
   // Implementation of runtime methods
   // =================================
@@ -148,6 +158,13 @@ class OptoRuntime : public AllStatic {
   static void multianewarray5_C(Klass* klass, int len1, int len2, int len3, int len4, int len5, JavaThread* current);
   static void multianewarrayN_C(Klass* klass, arrayOopDesc* dims, JavaThread* current);
 
+  // local methods passed as arguments to stub generator that forward
+  // control to corresponding JRT methods of SharedRuntime
+  static void slow_arraycopy_C(oopDesc* src,  jint src_pos,
+                               oopDesc* dest, jint dest_pos,
+                               jint length, JavaThread* thread);
+  static void complete_monitor_locking_C(oopDesc* obj, BasicLock* lock, JavaThread* current);
+
 public:
   static void monitor_notify_C(oopDesc* obj, JavaThread* current);
   static void monitor_notifyAll_C(oopDesc* obj, JavaThread* current);
@@ -168,13 +185,10 @@ class OptoRuntime : public AllStatic {
   // CodeBlob support
   // ===================================================================
 
-  static UncommonTrapBlob*   _uncommon_trap_blob;
-  static ExceptionBlob*       _exception_blob;
-
   static void generate_uncommon_trap_blob(void);
   static void generate_exception_blob();
 
-  static void register_finalizer(oopDesc* obj, JavaThread* current);
+  static void register_finalizer_C(oopDesc* obj, JavaThread* current);
 
  public:
 
@@ -188,6 +202,12 @@ class OptoRuntime : public AllStatic {
   // Returns the name of a stub
   static const char* stub_name(address entry);
 
+  // Returns the name associated with a given stub id
+  static const char* stub_name(OptoStubId id) {
+    assert(id > OptoStubId::NO_STUBID && id < OptoStubId::NUM_STUBIDS, "stub id out of range");
+    return _stub_names[(int)id];
+  }
+
   // access to runtime stubs entry points for java code
   static address new_instance_Java()                     { return _new_instance_Java; }
   static address new_array_Java()                        { return _new_array_Java; }
@@ -197,7 +217,6 @@ class OptoRuntime : public AllStatic {
   static address multianewarray4_Java()                  { return _multianewarray4_Java; }
   static address multianewarray5_Java()                  { return _multianewarray5_Java; }
   static address multianewarrayN_Java()                  { return _multianewarrayN_Java; }
-  static address vtable_must_compile_stub()              { return _vtable_must_compile_Java; }
   static address complete_monitor_locking_Java()         { return _complete_monitor_locking_Java; }
   static address monitor_notify_Java()                   { return _monitor_notify_Java; }
   static address monitor_notifyAll_Java()                { return _monitor_notifyAll_Java; }
@@ -227,6 +246,7 @@ class OptoRuntime : public AllStatic {
 
   static const TypeFunc* new_instance_Type(); // object allocation (slow case)
   static const TypeFunc* new_array_Type ();   // [a]newarray (slow case)
+  static const TypeFunc* new_array_nozero_Type ();   // [a]newarray (slow case)
   static const TypeFunc* multianewarray_Type(int ndim); // multianewarray
   static const TypeFunc* multianewarray2_Type(); // multianewarray
   static const TypeFunc* multianewarray3_Type(); // multianewarray
@@ -234,8 +254,10 @@ class OptoRuntime : public AllStatic {
   static const TypeFunc* multianewarray5_Type(); // multianewarray
   static const TypeFunc* multianewarrayN_Type(); // multianewarray
   static const TypeFunc* complete_monitor_enter_Type();
+  static const TypeFunc* complete_monitor_locking_Type();
   static const TypeFunc* complete_monitor_exit_Type();
   static const TypeFunc* monitor_notify_Type();
+  static const TypeFunc* monitor_notifyAll_Type();
   static const TypeFunc* uncommon_trap_Type();
   static const TypeFunc* athrow_Type();
   static const TypeFunc* rethrow_Type();
diff --git a/src/hotspot/share/runtime/stubDeclarations.hpp b/src/hotspot/share/runtime/stubDeclarations.hpp
index c9e946906f796..ccca14c61b5b8 100644
--- a/src/hotspot/share/runtime/stubDeclarations.hpp
+++ b/src/hotspot/share/runtime/stubDeclarations.hpp
@@ -70,13 +70,108 @@
   /* other stubs */                                                    \
   SHARED_JFR_STUBS_DO(do_blob)                                         \
 
-// generate a stub id enum tag from a name
+// C1 stubs are always generated in a generic CodeBlob
+
+#ifdef COMPILER1
+// do_blob(name)
+#define C1_STUBS_DO(do_blob)                                           \
+  do_blob(dtrace_object_alloc)                                         \
+  do_blob(unwind_exception)                                            \
+  do_blob(forward_exception)                                           \
+  do_blob(throw_range_check_failed)       /* throws ArrayIndexOutOfBoundsException */ \
+  do_blob(throw_index_exception)          /* throws IndexOutOfBoundsException */ \
+  do_blob(throw_div0_exception)                                        \
+  do_blob(throw_null_pointer_exception)                                \
+  do_blob(register_finalizer)                                          \
+  do_blob(new_instance)                                                \
+  do_blob(fast_new_instance)                                           \
+  do_blob(fast_new_instance_init_check)                                \
+  do_blob(new_type_array)                                              \
+  do_blob(new_object_array)                                            \
+  do_blob(new_multi_array)                                             \
+  do_blob(handle_exception_nofpu)         /* optimized version that does not preserve fpu registers */ \
+  do_blob(handle_exception)                                            \
+  do_blob(handle_exception_from_callee)                                \
+  do_blob(throw_array_store_exception)                                 \
+  do_blob(throw_class_cast_exception)                                  \
+  do_blob(throw_incompatible_class_change_error)                       \
+  do_blob(slow_subtype_check)                                          \
+  do_blob(monitorenter)                                                \
+  do_blob(monitorenter_nofpu)             /* optimized version that does not preserve fpu registers */ \
+  do_blob(monitorexit)                                                 \
+  do_blob(monitorexit_nofpu)              /* optimized version that does not preserve fpu registers */ \
+  do_blob(deoptimize)                                                  \
+  do_blob(access_field_patching)                                       \
+  do_blob(load_klass_patching)                                         \
+  do_blob(load_mirror_patching)                                        \
+  do_blob(load_appendix_patching)                                      \
+  do_blob(fpu2long_stub)                                               \
+  do_blob(counter_overflow)                                            \
+  do_blob(predicate_failed_trap)                                       \
+
+#else
+#define C1_STUBS_DO(do_blob)
+#endif
+
+// Opto stubs can be stored as entries with just an address or as
+// blobs of different types. The former may include some JVMTI stubs.
+//
+// n.b. blobs and stub defines are generated in the order defined by
+// C2_STUBS_DO, allowing dependencies from any givem stub on its
+// predecessors to be guaranteed. That explains the initial placement
+// of the blob declarations and intermediate placement of the jvmti
+// stubs.
+
+#ifdef COMPILER2
+// do_jvmti_stub(name)
+#if INCLUDE_JVMTI
+#define C2_JVMTI_STUBS_DO(do_jvmti_stub)                               \
+  do_jvmti_stub(notify_jvmti_vthread_start)                            \
+  do_jvmti_stub(notify_jvmti_vthread_end)                              \
+  do_jvmti_stub(notify_jvmti_vthread_mount)                            \
+  do_jvmti_stub(notify_jvmti_vthread_unmount)                          \
+
+#else
+#define C2_JVMTI_STUBS_DO(do_jvmti_stub)
+#endif // INCLUDE_JVMTI
+
+// do_blob(name, type)
+// do_stub(name, fancy_jump, pass_tls, return_pc)
+// do_jvmti_stub(name)
+//
+// n.b. non-jvmti stubs may employ a special type of jump (0, 1 or 2)
+// and require access to TLS and the return pc. jvmti stubs always
+// employ jump 0, and require no special access
+#define C2_STUBS_DO(do_blob, do_stub, do_jvmti_stub)                   \
+  do_blob(uncommon_trap, UncommonTrapBlob*)                            \
+  do_blob(exception, ExceptionBlob*)                                   \
+  do_stub(new_instance, 0, true, false)                                \
+  do_stub(new_array, 0, true, false)                                   \
+  do_stub(new_array_nozero, 0, true, false)                            \
+  do_stub(multianewarray2, 0, true, false)                             \
+  do_stub(multianewarray3, 0, true, false)                             \
+  do_stub(multianewarray4, 0, true, false)                             \
+  do_stub(multianewarray5, 0, true, false)                             \
+  do_stub(multianewarrayN, 0, true, false)                             \
+  C2_JVMTI_STUBS_DO(do_jvmti_stub)                                     \
+  do_stub(complete_monitor_locking, 0, false, false)                   \
+  do_stub(monitor_notify, 0, false, false)                             \
+  do_stub(monitor_notifyAll, 0, false, false)                          \
+  do_stub(rethrow, 2, true, true)                                      \
+  do_stub(slow_arraycopy, 0, false, false)                             \
+  do_stub(register_finalizer, 0, false, false)                         \
+
+#else
+#define C2_STUBS_DO(do_blob, do_stub, do_jvmti_stub)
+#endif
+
+// generate a stub or blob id enum tag from a name
 
 #define STUB_ID_NAME(base) base##_id
 
-// generate a blob id enum tag from a name
+// generate a stub field name
 
-#define BLOB_ID_NAME(base) base##_id
+#define STUB_FIELD_NAME(base) _##base
 
 // generate a blob field name
 
diff --git a/src/java.base/share/classes/java/lang/classfile/Annotation.java b/src/java.base/share/classes/java/lang/classfile/Annotation.java
index 357a7cb77ee83..009248ffd7856 100644
--- a/src/java.base/share/classes/java/lang/classfile/Annotation.java
+++ b/src/java.base/share/classes/java/lang/classfile/Annotation.java
@@ -34,6 +34,8 @@
 
 import java.lang.constant.ClassDesc;
 import java.util.List;
+
+import jdk.internal.classfile.impl.Util;
 import jdk.internal.javac.PreviewFeature;
 
 /**
@@ -78,7 +80,7 @@ public sealed interface Annotation
      * {@return the annotation interface, as a symbolic descriptor}
      */
     default ClassDesc classSymbol() {
-        return ClassDesc.ofDescriptor(className().stringValue());
+        return Util.fieldTypeSymbol(className());
     }
 
     /**
@@ -115,7 +117,7 @@ static Annotation of(Utf8Entry annotationClass,
      */
     static Annotation of(ClassDesc annotationClass,
                          List<AnnotationElement> elements) {
-        return of(TemporaryConstantPool.INSTANCE.utf8Entry(annotationClass.descriptorString()), elements);
+        return of(TemporaryConstantPool.INSTANCE.utf8Entry(annotationClass), elements);
     }
 
     /**
@@ -125,6 +127,6 @@ static Annotation of(ClassDesc annotationClass,
      */
     static Annotation of(ClassDesc annotationClass,
                          AnnotationElement... elements) {
-        return of(TemporaryConstantPool.INSTANCE.utf8Entry(annotationClass.descriptorString()), elements);
+        return of(TemporaryConstantPool.INSTANCE.utf8Entry(annotationClass), elements);
     }
 }
diff --git a/src/java.base/share/classes/java/lang/classfile/AnnotationValue.java b/src/java.base/share/classes/java/lang/classfile/AnnotationValue.java
index 613142077552d..c4474a248b520 100644
--- a/src/java.base/share/classes/java/lang/classfile/AnnotationValue.java
+++ b/src/java.base/share/classes/java/lang/classfile/AnnotationValue.java
@@ -38,6 +38,8 @@
 import java.lang.constant.Constable;
 import java.util.ArrayList;
 import java.util.List;
+
+import jdk.internal.classfile.impl.Util;
 import jdk.internal.javac.PreviewFeature;
 
 /**
@@ -405,7 +407,7 @@ sealed interface OfClass extends AnnotationValue
 
         /** {@return the class descriptor} */
         default ClassDesc classSymbol() {
-            return ClassDesc.ofDescriptor(className().stringValue());
+            return Util.fieldTypeSymbol(className());
         }
     }
 
@@ -423,7 +425,7 @@ sealed interface OfEnum extends AnnotationValue
 
         /** {@return the enum class descriptor} */
         default ClassDesc classSymbol() {
-            return ClassDesc.ofDescriptor(className().stringValue());
+            return Util.fieldTypeSymbol(className());
         }
 
         /** {@return the enum constant name} */
@@ -452,7 +454,7 @@ static OfEnum ofEnum(Utf8Entry className,
      * @param constantName the name of the enum constant
      */
     static OfEnum ofEnum(ClassDesc className, String constantName) {
-        return ofEnum(TemporaryConstantPool.INSTANCE.utf8Entry(className.descriptorString()),
+        return ofEnum(TemporaryConstantPool.INSTANCE.utf8Entry(className),
                       TemporaryConstantPool.INSTANCE.utf8Entry(constantName));
     }
 
@@ -469,7 +471,7 @@ static OfClass ofClass(Utf8Entry className) {
      * @param className the descriptor of the class
      */
     static OfClass ofClass(ClassDesc className) {
-        return ofClass(TemporaryConstantPool.INSTANCE.utf8Entry(className.descriptorString()));
+        return ofClass(TemporaryConstantPool.INSTANCE.utf8Entry(className));
     }
 
     /**
diff --git a/src/java.base/share/classes/java/lang/classfile/ClassBuilder.java b/src/java.base/share/classes/java/lang/classfile/ClassBuilder.java
index 1d5bb271dbe21..905c7355c3406 100644
--- a/src/java.base/share/classes/java/lang/classfile/ClassBuilder.java
+++ b/src/java.base/share/classes/java/lang/classfile/ClassBuilder.java
@@ -194,7 +194,9 @@ default ClassBuilder withField(String name,
     default ClassBuilder withField(String name,
                                    ClassDesc descriptor,
                                    int flags) {
-        return withField(name, descriptor, Util.buildingFlags(flags));
+        return withField(constantPool().utf8Entry(name),
+                         constantPool().utf8Entry(descriptor),
+                         flags);
     }
 
     /**
diff --git a/src/java.base/share/classes/java/lang/classfile/CodeBuilder.java b/src/java.base/share/classes/java/lang/classfile/CodeBuilder.java
index be0778d198458..5f4e5b72a1919 100644
--- a/src/java.base/share/classes/java/lang/classfile/CodeBuilder.java
+++ b/src/java.base/share/classes/java/lang/classfile/CodeBuilder.java
@@ -769,7 +769,7 @@ default CodeBuilder localVariable(int slot, Utf8Entry nameEntry, Utf8Entry descr
     default CodeBuilder localVariable(int slot, String name, ClassDesc descriptor, Label startScope, Label endScope) {
         return localVariable(slot,
                              constantPool().utf8Entry(name),
-                             constantPool().utf8Entry(descriptor.descriptorString()),
+                             constantPool().utf8Entry(descriptor),
                              startScope, endScope);
     }
 
diff --git a/src/java.base/share/classes/java/lang/classfile/FieldModel.java b/src/java.base/share/classes/java/lang/classfile/FieldModel.java
index e14f264ca2a98..006103d5f9cbc 100644
--- a/src/java.base/share/classes/java/lang/classfile/FieldModel.java
+++ b/src/java.base/share/classes/java/lang/classfile/FieldModel.java
@@ -31,6 +31,7 @@
 import java.lang.classfile.constantpool.Utf8Entry;
 import jdk.internal.classfile.impl.BufferedFieldBuilder;
 import jdk.internal.classfile.impl.FieldImpl;
+import jdk.internal.classfile.impl.Util;
 import jdk.internal.javac.PreviewFeature;
 
 /**
@@ -59,6 +60,6 @@ public sealed interface FieldModel
 
     /** {@return the field descriptor of this field, as a symbolic descriptor} */
     default ClassDesc fieldTypeSymbol() {
-        return ClassDesc.ofDescriptor(fieldType().stringValue());
+        return Util.fieldTypeSymbol(fieldType());
     }
 }
diff --git a/src/java.base/share/classes/java/lang/classfile/MethodModel.java b/src/java.base/share/classes/java/lang/classfile/MethodModel.java
index 651bc194ee3e0..bd51f3c97d72e 100644
--- a/src/java.base/share/classes/java/lang/classfile/MethodModel.java
+++ b/src/java.base/share/classes/java/lang/classfile/MethodModel.java
@@ -31,6 +31,7 @@
 import java.lang.classfile.constantpool.Utf8Entry;
 import jdk.internal.classfile.impl.BufferedMethodBuilder;
 import jdk.internal.classfile.impl.MethodImpl;
+import jdk.internal.classfile.impl.Util;
 import jdk.internal.javac.PreviewFeature;
 
 /**
@@ -59,7 +60,7 @@ public sealed interface MethodModel
 
     /** {@return the method descriptor of this method, as a symbolic descriptor} */
     default MethodTypeDesc methodTypeSymbol() {
-        return MethodTypeDesc.ofDescriptor(methodType().stringValue());
+        return Util.methodTypeSymbol(methodType());
     }
 
     /** {@return the body of this method, if there is one} */
diff --git a/src/java.base/share/classes/java/lang/classfile/attribute/LocalVariableInfo.java b/src/java.base/share/classes/java/lang/classfile/attribute/LocalVariableInfo.java
index 954682f665def..0ef8542e05c55 100644
--- a/src/java.base/share/classes/java/lang/classfile/attribute/LocalVariableInfo.java
+++ b/src/java.base/share/classes/java/lang/classfile/attribute/LocalVariableInfo.java
@@ -28,6 +28,7 @@
 import java.lang.classfile.constantpool.Utf8Entry;
 import jdk.internal.classfile.impl.BoundLocalVariable;
 import jdk.internal.classfile.impl.UnboundAttribute;
+import jdk.internal.classfile.impl.Util;
 import jdk.internal.javac.PreviewFeature;
 
 /**
@@ -65,7 +66,7 @@ public sealed interface LocalVariableInfo
      * {@return the field descriptor of the local variable}
      */
     default ClassDesc typeSymbol() {
-        return ClassDesc.ofDescriptor(type().stringValue());
+        return Util.fieldTypeSymbol(type());
     }
 
     /**
diff --git a/src/java.base/share/classes/java/lang/classfile/attribute/RecordComponentInfo.java b/src/java.base/share/classes/java/lang/classfile/attribute/RecordComponentInfo.java
index ab7df7a5db2b4..5a4c0d87b839a 100644
--- a/src/java.base/share/classes/java/lang/classfile/attribute/RecordComponentInfo.java
+++ b/src/java.base/share/classes/java/lang/classfile/attribute/RecordComponentInfo.java
@@ -33,6 +33,7 @@
 import jdk.internal.classfile.impl.BoundRecordComponentInfo;
 import jdk.internal.classfile.impl.TemporaryConstantPool;
 import jdk.internal.classfile.impl.UnboundAttribute;
+import jdk.internal.classfile.impl.Util;
 import jdk.internal.javac.PreviewFeature;
 
 /**
@@ -58,7 +59,7 @@ public sealed interface RecordComponentInfo
      * {@return the field descriptor of this component, as a {@linkplain ClassDesc}}
      */
     default ClassDesc descriptorSymbol() {
-        return ClassDesc.ofDescriptor(descriptor().stringValue());
+        return Util.fieldTypeSymbol(descriptor());
     }
 
     /**
@@ -95,7 +96,7 @@ static RecordComponentInfo of(String name,
                                   ClassDesc descriptor,
                                   List<Attribute<?>> attributes) {
         return new UnboundAttribute.UnboundRecordComponentInfo(TemporaryConstantPool.INSTANCE.utf8Entry(name),
-                                                               TemporaryConstantPool.INSTANCE.utf8Entry(descriptor.descriptorString()),
+                                                               TemporaryConstantPool.INSTANCE.utf8Entry(descriptor),
                                                                attributes);
     }
 
diff --git a/src/java.base/share/classes/java/lang/classfile/constantpool/ConstantPoolBuilder.java b/src/java.base/share/classes/java/lang/classfile/constantpool/ConstantPoolBuilder.java
index 7334d8e54604e..1c0d6e55e3143 100644
--- a/src/java.base/share/classes/java/lang/classfile/constantpool/ConstantPoolBuilder.java
+++ b/src/java.base/share/classes/java/lang/classfile/constantpool/ConstantPoolBuilder.java
@@ -223,9 +223,7 @@ default ModuleEntry moduleEntry(ModuleDesc moduleDesc) {
      * @param type the symbolic descriptor for a field type
      */
     default NameAndTypeEntry nameAndTypeEntry(String name, ClassDesc type) {
-        var ret = (NameAndTypeEntryImpl)nameAndTypeEntry(utf8Entry(name), utf8Entry(type.descriptorString()));
-        ret.typeSym = type;
-        return ret;
+        return nameAndTypeEntry(utf8Entry(name), utf8Entry(type));
     }
 
     /**
@@ -238,9 +236,7 @@ default NameAndTypeEntry nameAndTypeEntry(String name, ClassDesc type) {
      * @param type the symbolic descriptor for a method type
      */
     default NameAndTypeEntry nameAndTypeEntry(String name, MethodTypeDesc type) {
-        var ret = (NameAndTypeEntryImpl)nameAndTypeEntry(utf8Entry(name), utf8Entry(type.descriptorString()));
-        ret.typeSym = type;
-        return ret;
+        return nameAndTypeEntry(utf8Entry(name), utf8Entry(type));
     }
 
     /**
diff --git a/src/java.base/share/classes/java/lang/classfile/instruction/LocalVariable.java b/src/java.base/share/classes/java/lang/classfile/instruction/LocalVariable.java
index 4d3d755386763..84a8599315218 100644
--- a/src/java.base/share/classes/java/lang/classfile/instruction/LocalVariable.java
+++ b/src/java.base/share/classes/java/lang/classfile/instruction/LocalVariable.java
@@ -36,6 +36,7 @@
 import jdk.internal.classfile.impl.AbstractPseudoInstruction;
 import jdk.internal.classfile.impl.BoundLocalVariable;
 import jdk.internal.classfile.impl.TemporaryConstantPool;
+import jdk.internal.classfile.impl.Util;
 import jdk.internal.javac.PreviewFeature;
 
 /**
@@ -70,7 +71,7 @@ public sealed interface LocalVariable extends PseudoInstruction
      * {@return the local variable type, as a symbolic descriptor}
      */
     default ClassDesc typeSymbol() {
-        return ClassDesc.ofDescriptor(type().stringValue());
+        return Util.fieldTypeSymbol(type());
     }
 
     /**
@@ -109,7 +110,7 @@ static LocalVariable of(int slot, Utf8Entry nameEntry, Utf8Entry descriptorEntry
     static LocalVariable of(int slot, String name, ClassDesc descriptor, Label startScope, Label endScope) {
         return of(slot,
                   TemporaryConstantPool.INSTANCE.utf8Entry(name),
-                  TemporaryConstantPool.INSTANCE.utf8Entry(descriptor.descriptorString()),
+                  TemporaryConstantPool.INSTANCE.utf8Entry(descriptor),
                   startScope, endScope);
     }
 }
diff --git a/src/java.base/share/classes/java/lang/foreign/MemorySegment.java b/src/java.base/share/classes/java/lang/foreign/MemorySegment.java
index 38fd36bbb1584..cb1f3707db68c 100644
--- a/src/java.base/share/classes/java/lang/foreign/MemorySegment.java
+++ b/src/java.base/share/classes/java/lang/foreign/MemorySegment.java
@@ -43,6 +43,7 @@
 import java.util.stream.Stream;
 import jdk.internal.foreign.AbstractMemorySegmentImpl;
 import jdk.internal.foreign.MemorySessionImpl;
+import jdk.internal.foreign.SegmentBulkOperations;
 import jdk.internal.foreign.SegmentFactories;
 import jdk.internal.javac.Restricted;
 import jdk.internal.reflect.CallerSensitive;
@@ -1571,7 +1572,7 @@ static MemorySegment ofAddress(long address) {
     static void copy(MemorySegment srcSegment, long srcOffset,
                      MemorySegment dstSegment, long dstOffset, long bytes) {
 
-        AbstractMemorySegmentImpl.copy((AbstractMemorySegmentImpl) srcSegment, srcOffset,
+        SegmentBulkOperations.copy((AbstractMemorySegmentImpl) srcSegment, srcOffset,
                 (AbstractMemorySegmentImpl) dstSegment, dstOffset,
                 bytes);
     }
@@ -2635,8 +2636,9 @@ static void copy(Object srcArray, int srcIndex,
      */
     static long mismatch(MemorySegment srcSegment, long srcFromOffset, long srcToOffset,
                          MemorySegment dstSegment, long dstFromOffset, long dstToOffset) {
-        return AbstractMemorySegmentImpl.mismatch(srcSegment, srcFromOffset, srcToOffset,
-                dstSegment, dstFromOffset, dstToOffset);
+        return SegmentBulkOperations.mismatch(
+                (AbstractMemorySegmentImpl)Objects.requireNonNull(srcSegment), srcFromOffset, srcToOffset,
+                (AbstractMemorySegmentImpl)Objects.requireNonNull(dstSegment), dstFromOffset, dstToOffset);
     }
 
     /**
diff --git a/src/java.base/share/classes/java/util/zip/ZipCoder.java b/src/java.base/share/classes/java/util/zip/ZipCoder.java
index 25c6e3a2e4acd..6692703c1b3f0 100644
--- a/src/java.base/share/classes/java/util/zip/ZipCoder.java
+++ b/src/java.base/share/classes/java/util/zip/ZipCoder.java
@@ -158,13 +158,6 @@ static int hash(String name) {
         return hsh;
     }
 
-    boolean hasTrailingSlash(byte[] a, int end) {
-        byte[] slashBytes = slashBytes();
-        return end >= slashBytes.length &&
-            Arrays.mismatch(a, end - slashBytes.length, end, slashBytes, 0, slashBytes.length) == -1;
-    }
-
-    private byte[] slashBytes;
     private final Charset cs;
     protected CharsetDecoder dec;
     private CharsetEncoder enc;
@@ -191,23 +184,6 @@ private CharsetEncoder encoder() {
         return enc;
     }
 
-    // This method produces an array with the bytes that will correspond to a
-    // trailing '/' in the chosen character encoding.
-    //
-    // While in most charsets a trailing slash will be encoded as the byte
-    // value of '/', this does not hold in the general case. E.g., in charsets
-    // such as UTF-16 and UTF-32 it will be represented by a sequence of 2 or 4
-    // bytes, respectively.
-    private byte[] slashBytes() {
-        if (slashBytes == null) {
-            // Take into account charsets that produce a BOM, e.g., UTF-16
-            byte[] slash = "/".getBytes(cs);
-            byte[] doubleSlash = "//".getBytes(cs);
-            slashBytes = Arrays.copyOfRange(doubleSlash, slash.length, doubleSlash.length);
-        }
-        return slashBytes;
-    }
-
     /**
      * This method is used by ZipFile.Source.getEntryPos when comparing the
      * name being looked up to candidate names encoded in the CEN byte
@@ -297,8 +273,7 @@ int checkedHash(byte[] a, int off, int len) throws Exception {
             return h;
         }
 
-        @Override
-        boolean hasTrailingSlash(byte[] a, int end) {
+        private boolean hasTrailingSlash(byte[] a, int end) {
             return end > 0 && a[end - 1] == '/';
         }
 
diff --git a/src/java.base/share/classes/java/util/zip/ZipFile.java b/src/java.base/share/classes/java/util/zip/ZipFile.java
index 6ab70d4bdb81d..80126bc79734d 100644
--- a/src/java.base/share/classes/java/util/zip/ZipFile.java
+++ b/src/java.base/share/classes/java/util/zip/ZipFile.java
@@ -347,9 +347,12 @@ public ZipEntry getEntry(String name) {
         ZipEntry entry = null;
         synchronized (this) {
             ensureOpen();
-            int pos = res.zsrc.getEntryPos(name, true);
-            if (pos != -1) {
-                entry = getZipEntry(name, pos);
+            // Look up the name and CEN header position of the entry.
+            // The resolved name may include a trailing slash.
+            // See Source::getEntryPos for details.
+            EntryPos pos = res.zsrc.getEntryPos(name, true);
+            if (pos != null) {
+                entry = getZipEntry(pos.name, pos.pos);
             }
         }
         return entry;
@@ -387,7 +390,12 @@ public InputStream getInputStream(ZipEntry entry) throws IOException {
             if (Objects.equals(lastEntryName, entry.name)) {
                 pos = lastEntryPos;
             } else {
-                pos = zsrc.getEntryPos(entry.name, false);
+                EntryPos entryPos = zsrc.getEntryPos(entry.name, false);
+                if (entryPos != null) {
+                    pos = entryPos.pos;
+                } else {
+                    pos = -1;
+                }
             }
             if (pos == -1) {
                 return null;
@@ -540,7 +548,8 @@ public T next() {
                     throw new NoSuchElementException();
                 }
                 // each "entry" has 3 ints in table entries
-                return (T)getZipEntry(null, res.zsrc.getEntryPos(i++ * 3));
+                int pos = res.zsrc.getEntryPos(i++ * 3);
+                return (T)getZipEntry(getEntryName(pos), pos);
             }
         }
 
@@ -612,7 +621,7 @@ public Stream<? extends ZipEntry> stream() {
         synchronized (this) {
             ensureOpen();
             return StreamSupport.stream(new EntrySpliterator<>(0, res.zsrc.total,
-                pos -> getZipEntry(null, pos)), false);
+                pos -> getZipEntry(getEntryName(pos), pos)), false);
        }
     }
 
@@ -655,7 +664,7 @@ private Stream<JarEntry> jarStream() {
         synchronized (this) {
             ensureOpen();
             return StreamSupport.stream(new EntrySpliterator<>(0, res.zsrc.total,
-                pos -> (JarEntry)getZipEntry(null, pos)), false);
+                pos -> (JarEntry)getZipEntry(getEntryName(pos), pos)), false);
         }
     }
 
@@ -665,30 +674,10 @@ private Stream<JarEntry> jarStream() {
     /* Check ensureOpen() before invoking this method */
     private ZipEntry getZipEntry(String name, int pos) {
         byte[] cen = res.zsrc.cen;
-        int nlen = CENNAM(cen, pos);
-        int elen = CENEXT(cen, pos);
-        int clen = CENCOM(cen, pos);
+        ZipEntry e = this instanceof JarFile jarFile
+                ? Source.JUJA.entryFor(jarFile, name)
+                : new ZipEntry(name);
 
-        ZipCoder zc = res.zsrc.zipCoderForPos(pos);
-        if (name != null) {
-            // only need to check for mismatch of trailing slash
-            if (nlen > 0 &&
-                !name.isEmpty() &&
-                zc.hasTrailingSlash(cen, pos + CENHDR + nlen) &&
-                !name.endsWith("/"))
-            {
-                name += '/';
-            }
-        } else {
-            // invoked from iterator, use the entry name stored in cen
-            name = zc.toString(cen, pos + CENHDR, nlen);
-        }
-        ZipEntry e;
-        if (this instanceof JarFile) {
-            e = Source.JUJA.entryFor((JarFile)this, name);
-        } else {
-            e = new ZipEntry(name);
-        }
         e.flag = CENFLG(cen, pos);
         e.xdostime = CENTIM(cen, pos);
         e.crc = CENCRC(cen, pos);
@@ -700,12 +689,17 @@ private ZipEntry getZipEntry(String name, int pos) {
             e.externalFileAttributes = CENATX_PERMS(cen, pos) & 0xFFFF;
         }
 
+        int nlen = CENNAM(cen, pos);
+        int elen = CENEXT(cen, pos);
+        int clen = CENCOM(cen, pos);
+
         if (elen != 0) {
             int start = pos + CENHDR + nlen;
             e.setExtra0(Arrays.copyOfRange(cen, start, start + elen), true, false);
         }
         if (clen != 0) {
             int start = pos + CENHDR + nlen + elen;
+            ZipCoder zc = res.zsrc.zipCoderForPos(pos);
             e.comment = zc.toString(cen, start, clen);
         }
         lastEntryName = e.name;
@@ -1176,6 +1170,8 @@ public void setExternalFileAttributes(ZipEntry ze, int externalFileAttributes) {
              }
         );
     }
+    // Represents the resolved name and position of a CEN record
+    static record EntryPos(String name, int pos) {}
 
     private static class Source {
         // While this is only used from ZipFile, defining it there would cause
@@ -1849,12 +1845,12 @@ private static void zerror(String msg) throws ZipException {
         }
 
         /*
-         * Returns the {@code pos} of the ZIP cen entry corresponding to the
-         * specified entry name, or -1 if not found.
+         * Returns the resolved name and position of the ZIP cen entry corresponding
+         *  to the specified entry name, or {@code null} if not found.
          */
-        private int getEntryPos(String name, boolean addSlash) {
+        private EntryPos getEntryPos(String name, boolean addSlash) {
             if (total == 0) {
-                return -1;
+                return null;
             }
 
             int hsh = ZipCoder.hash(name);
@@ -1877,7 +1873,7 @@ private int getEntryPos(String name, boolean addSlash) {
                     switch (zc.compare(name, cen, noff, nlen, addSlash)) {
                         case EXACT_MATCH:
                             // We found an exact match for "name"
-                            return pos;
+                            return new EntryPos(name, pos);
                         case DIRECTORY_MATCH:
                             // We found the directory "name/"
                             // Track its position, then continue the search for "name"
@@ -1892,10 +1888,10 @@ private int getEntryPos(String name, boolean addSlash) {
             // Reaching this point means we did not find "name".
             // Return the position of "name/" if we found it
             if (dirPos != -1) {
-                return dirPos;
+                return new EntryPos(name + "/", dirPos);
             }
             // No entry found
-            return -1;
+            return null;
         }
 
         private ZipCoder zipCoderForPos(int pos) {
diff --git a/src/java.base/share/classes/jdk/internal/classfile/impl/AbstractPoolEntry.java b/src/java.base/share/classes/jdk/internal/classfile/impl/AbstractPoolEntry.java
index 450f6ae1c8c90..7f14560fd1854 100644
--- a/src/java.base/share/classes/jdk/internal/classfile/impl/AbstractPoolEntry.java
+++ b/src/java.base/share/classes/jdk/internal/classfile/impl/AbstractPoolEntry.java
@@ -54,6 +54,7 @@
 import jdk.internal.access.JavaLangAccess;
 import jdk.internal.access.SharedSecrets;
 import jdk.internal.util.ArraysSupport;
+import jdk.internal.vm.annotation.Stable;
 
 public abstract sealed class AbstractPoolEntry {
     /*
@@ -146,12 +147,14 @@ enum State { RAW, BYTE, CHAR, STRING }
         private final int offset;
         private final int rawLen;
         // Set in any state other than RAW
-        private int hash;
-        private int charLen;
+        private @Stable int hash;
+        private @Stable int charLen;
         // Set in CHAR state
-        private char[] chars;
+        private @Stable char[] chars;
         // Only set in STRING state
-        private String stringValue;
+        private @Stable String stringValue;
+        // The descriptor symbol, if this is a descriptor
+        @Stable TypeDescriptor typeSym;
 
         Utf8EntryImpl(ConstantPool cpm, int index,
                           byte[] rawBytes, int offset, int rawLen) {
@@ -187,6 +190,7 @@ enum State { RAW, BYTE, CHAR, STRING }
             this.charLen = u.charLen;
             this.chars = u.chars;
             this.stringValue = u.stringValue;
+            this.typeSym = u.typeSym;
         }
 
         /**
@@ -419,6 +423,22 @@ void writeTo(BufWriterImpl pool) {
                 pool.writeUTF(stringValue);
             }
         }
+
+        public ClassDesc fieldTypeSymbol() {
+            if (typeSym instanceof ClassDesc cd)
+                return cd;
+            var ret = ClassDesc.ofDescriptor(stringValue());
+            typeSym = ret;
+            return ret;
+        }
+
+        public MethodTypeDesc methodTypeSymbol() {
+            if (typeSym instanceof MethodTypeDesc mtd)
+                return mtd;
+            var ret = MethodTypeDesc.ofDescriptor(stringValue());
+            typeSym = ret;
+            return ret;
+        }
     }
 
     abstract static sealed class AbstractRefEntry<T extends PoolEntry> extends AbstractPoolEntry {
@@ -585,8 +605,6 @@ public boolean equals(Object o) {
     public static final class NameAndTypeEntryImpl extends AbstractRefsEntry<Utf8EntryImpl, Utf8EntryImpl>
             implements NameAndTypeEntry {
 
-        public TypeDescriptor typeSym = null;
-
         NameAndTypeEntryImpl(ConstantPool cpm, int index, Utf8EntryImpl name, Utf8EntryImpl type) {
             super(cpm, ClassFile.TAG_NAMEANDTYPE, index, name, type);
         }
@@ -601,31 +619,12 @@ public Utf8Entry type() {
             return ref2;
         }
 
-        public ClassDesc fieldTypeSymbol() {
-            if (typeSym instanceof ClassDesc cd) {
-                return cd;
-            } else {
-                return (ClassDesc)(typeSym = ClassDesc.ofDescriptor(ref2.stringValue()));
-            }
-        }
-
-        public MethodTypeDesc methodTypeSymbol() {
-            if (typeSym instanceof MethodTypeDesc mtd) {
-                return mtd;
-            } else {
-                return (MethodTypeDesc)(typeSym = MethodTypeDesc.ofDescriptor(ref2.stringValue()));
-            }
-        }
-
         @Override
         public NameAndTypeEntry clone(ConstantPoolBuilder cp) {
             if (cp.canWriteDirect(constantPool)) {
                 return this;
-            } else {
-                var ret = (NameAndTypeEntryImpl)cp.nameAndTypeEntry(ref1, ref2);
-                ret.typeSym = typeSym;
-                return ret;
             }
+            return cp.nameAndTypeEntry(ref1, ref2);
         }
 
         @Override
@@ -897,8 +896,6 @@ public static final class MethodTypeEntryImpl
             extends AbstractRefEntry<Utf8EntryImpl>
             implements MethodTypeEntry {
 
-        public MethodTypeDesc sym = null;
-
         MethodTypeEntryImpl(ConstantPool cpm, int index, Utf8EntryImpl descriptor) {
             super(cpm, ClassFile.TAG_METHODTYPE, index, descriptor);
         }
@@ -912,20 +909,13 @@ public Utf8Entry descriptor() {
         public MethodTypeEntry clone(ConstantPoolBuilder cp) {
             if (cp.canWriteDirect(constantPool)) {
                 return this;
-            } else {
-                var ret = (MethodTypeEntryImpl)cp.methodTypeEntry(ref1);
-                ret.sym = sym;
-                return ret;
             }
+            return cp.methodTypeEntry(ref1);
         }
 
         @Override
         public MethodTypeDesc asSymbol() {
-            var sym = this.sym;
-            if (sym != null) {
-                return sym;
-            }
-            return this.sym = MethodTypeDesc.ofDescriptor(descriptor().stringValue());
+            return ref1.methodTypeSymbol();
         }
 
         @Override
diff --git a/src/java.base/share/classes/jdk/internal/classfile/impl/BoundLocalVariable.java b/src/java.base/share/classes/jdk/internal/classfile/impl/BoundLocalVariable.java
index a5953c86f5dcc..3bbcd243cc11e 100644
--- a/src/java.base/share/classes/jdk/internal/classfile/impl/BoundLocalVariable.java
+++ b/src/java.base/share/classes/jdk/internal/classfile/impl/BoundLocalVariable.java
@@ -46,7 +46,7 @@ public Utf8Entry type() {
 
     @Override
     public ClassDesc typeSymbol() {
-        return ClassDesc.ofDescriptor(type().stringValue());
+        return Util.fieldTypeSymbol(type());
     }
 
     @Override
diff --git a/src/java.base/share/classes/jdk/internal/classfile/impl/BufferedMethodBuilder.java b/src/java.base/share/classes/jdk/internal/classfile/impl/BufferedMethodBuilder.java
index 84ddd09b990ec..bc6ab555ae597 100644
--- a/src/java.base/share/classes/jdk/internal/classfile/impl/BufferedMethodBuilder.java
+++ b/src/java.base/share/classes/jdk/internal/classfile/impl/BufferedMethodBuilder.java
@@ -53,7 +53,6 @@ public final class BufferedMethodBuilder
     private AccessFlags flags;
     private final MethodModel original;
     private int[] parameterSlots;
-    MethodTypeDesc mDesc;
 
     public BufferedMethodBuilder(SplitConstantPool constantPool,
                                  ClassFileImpl context,
@@ -102,14 +101,7 @@ public Utf8Entry methodType() {
 
     @Override
     public MethodTypeDesc methodTypeSymbol() {
-        if (mDesc == null) {
-            if (original instanceof MethodInfo mi) {
-                mDesc = mi.methodTypeSymbol();
-            } else {
-                mDesc = MethodTypeDesc.ofDescriptor(methodType().stringValue());
-            }
-        }
-        return mDesc;
+        return Util.methodTypeSymbol(methodType());
     }
 
     @Override
diff --git a/src/java.base/share/classes/jdk/internal/classfile/impl/ChainedClassBuilder.java b/src/java.base/share/classes/jdk/internal/classfile/impl/ChainedClassBuilder.java
index 50c1590e8a271..8f05f20d739be 100644
--- a/src/java.base/share/classes/jdk/internal/classfile/impl/ChainedClassBuilder.java
+++ b/src/java.base/share/classes/jdk/internal/classfile/impl/ChainedClassBuilder.java
@@ -79,15 +79,6 @@ public ClassBuilder withMethod(Utf8Entry name, Utf8Entry descriptor, int flags,
         return this;
     }
 
-    @Override
-    public ClassBuilder withMethod(String name, MethodTypeDesc descriptor, int flags, Consumer<? super MethodBuilder> handler) {
-        var mb = new BufferedMethodBuilder(terminal.constantPool, terminal.context,
-                constantPool().utf8Entry(name), constantPool().utf8Entry(descriptor), flags, null);
-        mb.mDesc = descriptor;
-        consumer.accept(mb.run(handler).toModel());
-        return this;
-    }
-
     @Override
     public ClassBuilder transformMethod(MethodModel method, MethodTransform transform) {
         BufferedMethodBuilder builder = new BufferedMethodBuilder(terminal.constantPool, terminal.context,
diff --git a/src/java.base/share/classes/jdk/internal/classfile/impl/DirectClassBuilder.java b/src/java.base/share/classes/jdk/internal/classfile/impl/DirectClassBuilder.java
index 7d67ed272aff2..6901ae7e24cf9 100644
--- a/src/java.base/share/classes/jdk/internal/classfile/impl/DirectClassBuilder.java
+++ b/src/java.base/share/classes/jdk/internal/classfile/impl/DirectClassBuilder.java
@@ -90,14 +90,6 @@ public ClassBuilder withFlags(int flags) {
         return this;
     }
 
-    @Override
-    public ClassBuilder withField(String name,
-                                  ClassDesc descriptor,
-                                  int flags) {
-        return withField(new DirectFieldBuilder(constantPool, context,
-                constantPool.utf8Entry(name), constantPool.utf8Entry(descriptor), flags, null));
-    }
-
     @Override
     public ClassBuilder withField(Utf8Entry name,
                                   Utf8Entry descriptor,
@@ -130,13 +122,6 @@ public ClassBuilder withMethod(Utf8Entry name,
                                   .run(handler));
     }
 
-    @Override
-    public ClassBuilder withMethod(String name, MethodTypeDesc descriptor, int flags, Consumer<? super MethodBuilder> handler) {
-        var method = new DirectMethodBuilder(constantPool, context, constantPool.utf8Entry(name), constantPool.utf8Entry(descriptor), flags, null);
-        method.mDesc = descriptor;
-        return withMethod(method.run(handler));
-    }
-
     @Override
     public ClassBuilder transformMethod(MethodModel method, MethodTransform transform) {
         DirectMethodBuilder builder = new DirectMethodBuilder(constantPool, context, method.methodName(),
diff --git a/src/java.base/share/classes/jdk/internal/classfile/impl/DirectMethodBuilder.java b/src/java.base/share/classes/jdk/internal/classfile/impl/DirectMethodBuilder.java
index fac6b7384e2fc..ec4e148baf30d 100644
--- a/src/java.base/share/classes/jdk/internal/classfile/impl/DirectMethodBuilder.java
+++ b/src/java.base/share/classes/jdk/internal/classfile/impl/DirectMethodBuilder.java
@@ -46,7 +46,6 @@ public final class DirectMethodBuilder
     final Utf8Entry desc;
     int flags;
     int[] parameterSlots;
-    MethodTypeDesc mDesc;
 
     public DirectMethodBuilder(SplitConstantPool constantPool,
                                ClassFileImpl context,
@@ -87,14 +86,7 @@ public Utf8Entry methodType() {
 
     @Override
     public MethodTypeDesc methodTypeSymbol() {
-        if (mDesc == null) {
-            if (original instanceof MethodInfo mi) {
-                mDesc = mi.methodTypeSymbol();
-            } else {
-                mDesc = MethodTypeDesc.ofDescriptor(methodType().stringValue());
-            }
-        }
-        return mDesc;
+        return Util.methodTypeSymbol(methodType());
     }
 
     @Override
diff --git a/src/java.base/share/classes/jdk/internal/classfile/impl/EntryMap.java b/src/java.base/share/classes/jdk/internal/classfile/impl/EntryMap.java
index 4270dbf79a36f..d5ebbbebe7ada 100644
--- a/src/java.base/share/classes/jdk/internal/classfile/impl/EntryMap.java
+++ b/src/java.base/share/classes/jdk/internal/classfile/impl/EntryMap.java
@@ -33,10 +33,10 @@
  * element is the hash and the second is the mapped index.  To look something up
  * in the map, provide a hash value and an index to map it to, and invoke
  * firstToken(hash).  This returns an opaque token that can be provided to
- * nextToken(hash, token) to get the next candidate, or to getElementByToken(token)
- * or getIndexByToken to get the mapped element or index.
+ * nextToken(hash, token) to get the next candidate, or to getIndexByToken to
+ * get the mapped element or index.
  */
-public abstract class EntryMap<T> {
+public final class EntryMap {
     public static final int NO_VALUE = -1;
 
     /**
@@ -77,8 +77,6 @@ public EntryMap(int size, float fillFactor) {
         data = new int[capacity * 2];
     }
 
-    protected abstract T fetchElement(int index);
-
     public int firstToken(int hash) {
         if (hash == 0)
             throw new IllegalArgumentException("hash must be nonzero");
@@ -110,10 +108,6 @@ public int getIndexByToken(int token) {
         return data[token + 1];
     }
 
-    public T getElementByToken(int token) {
-        return fetchElement(data[token + 1]);
-    }
-
     public void put(int hash, int index) {
         if (hash == 0)
             throw new IllegalArgumentException("hash must be nonzero");
diff --git a/src/java.base/share/classes/jdk/internal/classfile/impl/MethodImpl.java b/src/java.base/share/classes/jdk/internal/classfile/impl/MethodImpl.java
index 05de881ba297f..8467152504ee5 100644
--- a/src/java.base/share/classes/jdk/internal/classfile/impl/MethodImpl.java
+++ b/src/java.base/share/classes/jdk/internal/classfile/impl/MethodImpl.java
@@ -41,7 +41,6 @@ public final class MethodImpl
     private final int startPos, endPos, attributesPos;
     private List<Attribute<?>> attributes;
     private int[] parameterSlots;
-    private MethodTypeDesc mDesc;
 
     public MethodImpl(ClassReader reader, int startPos, int endPos, int attrStart) {
         this.reader = reader;
@@ -75,10 +74,7 @@ public Utf8Entry methodType() {
 
     @Override
     public MethodTypeDesc methodTypeSymbol() {
-        if (mDesc == null) {
-            mDesc = MethodTypeDesc.ofDescriptor(methodType().stringValue());
-        }
-        return mDesc;
+        return Util.methodTypeSymbol(methodType());
     }
 
     @Override
diff --git a/src/java.base/share/classes/jdk/internal/classfile/impl/SplitConstantPool.java b/src/java.base/share/classes/jdk/internal/classfile/impl/SplitConstantPool.java
index 4f48c8b731db0..0b5d64023b962 100644
--- a/src/java.base/share/classes/jdk/internal/classfile/impl/SplitConstantPool.java
+++ b/src/java.base/share/classes/jdk/internal/classfile/impl/SplitConstantPool.java
@@ -24,7 +24,7 @@
  */
 package jdk.internal.classfile.impl;
 
-import java.lang.constant.ConstantDesc;
+import java.lang.constant.ClassDesc;
 import java.lang.constant.MethodTypeDesc;
 import java.util.Arrays;
 import java.util.List;
@@ -63,8 +63,8 @@ public final class SplitConstantPool implements ConstantPoolBuilder {
     private PoolEntry[] myEntries;
     private BootstrapMethodEntryImpl[] myBsmEntries;
     private boolean doneFullScan;
-    private EntryMap<PoolEntry> map;
-    private EntryMap<BootstrapMethodEntryImpl> bsmMap;
+    private EntryMap map;
+    private EntryMap bsmMap;
 
     public SplitConstantPool() {
         this.size = 1;
@@ -178,14 +178,10 @@ void writeTo(BufWriterImpl buf) {
         }
     }
 
-    private EntryMap<PoolEntry> map() {
+    private EntryMap map() {
         if (map == null) {
-            map = new EntryMap<>(Math.max(size, 1024), .75f) {
-                @Override
-                protected PoolEntry fetchElement(int index) {
-                    return entryByIndex(index);
-                }
-            };
+            map = new EntryMap(Math.max(size, 1024), .75f);
+
             // Doing a full scan here yields fall-off-the-cliff performance results,
             // especially if we only need a few entries that are already
             // inflated (such as attribute names).
@@ -220,14 +216,9 @@ private void fullScan() {
         doneFullScan = true;
     }
 
-    private EntryMap<BootstrapMethodEntryImpl> bsmMap() {
+    private EntryMap bsmMap() {
         if (bsmMap == null) {
-            bsmMap = new EntryMap<>(Math.max(bsmSize, 16), .75f) {
-                @Override
-                protected BootstrapMethodEntryImpl fetchElement(int index) {
-                    return bootstrapMethodEntry(index);
-                }
-            };
+            bsmMap = new EntryMap(Math.max(bsmSize, 16), .75f);
             for (int i=0; i<parentBsmSize; i++) {
                 BootstrapMethodEntryImpl bsm = parent.bootstrapMethodEntry(i);
                 bsmMap.put(bsm.hash, bsm.index);
@@ -268,9 +259,9 @@ private BootstrapMethodEntryImpl internalAdd(BootstrapMethodEntryImpl bsm, int h
 
     private IntegerEntry findIntEntry(int val) {
         int hash = AbstractPoolEntry.hash1(TAG_INTEGER, Integer.hashCode(val));
-        EntryMap<PoolEntry> map = map();
+        EntryMap map = map();
         for (int token = map.firstToken(hash); token != -1; token = map.nextToken(hash, token)) {
-            PoolEntry e = map.getElementByToken(token);
+            PoolEntry e = entryByIndex(map.getIndexByToken(token));
             if (e.tag() == TAG_INTEGER
                     && e instanceof AbstractPoolEntry.IntegerEntryImpl ce
                     && ce.intValue() == val)
@@ -285,9 +276,9 @@ private IntegerEntry findIntEntry(int val) {
 
     private LongEntry findLongEntry(long val) {
         int hash = AbstractPoolEntry.hash1(TAG_LONG, Long.hashCode(val));
-        EntryMap<PoolEntry> map = map();
+        EntryMap map = map();
         for (int token = map.firstToken(hash); token != -1; token = map.nextToken(hash, token)) {
-            PoolEntry e = map.getElementByToken(token);
+            PoolEntry e = entryByIndex(map.getIndexByToken(token));
             if (e.tag() == TAG_LONG
                     && e instanceof AbstractPoolEntry.LongEntryImpl ce
                     && ce.longValue() == val)
@@ -302,9 +293,9 @@ private LongEntry findLongEntry(long val) {
 
     private FloatEntry findFloatEntry(float val) {
         int hash = AbstractPoolEntry.hash1(TAG_FLOAT, Float.hashCode(val));
-        EntryMap<PoolEntry> map = map();
+        EntryMap map = map();
         for (int token = map.firstToken(hash); token != -1; token = map.nextToken(hash, token)) {
-            PoolEntry e = map.getElementByToken(token);
+            PoolEntry e = entryByIndex(map.getIndexByToken(token));
             if (e.tag() == TAG_FLOAT
                     && e instanceof AbstractPoolEntry.FloatEntryImpl ce
                     && ce.floatValue() == val)
@@ -319,9 +310,9 @@ private FloatEntry findFloatEntry(float val) {
 
     private DoubleEntry findDoubleEntry(double val) {
         int hash = AbstractPoolEntry.hash1(TAG_DOUBLE, Double.hashCode(val));
-        EntryMap<PoolEntry> map = map();
+        EntryMap map = map();
         for (int token = map.firstToken(hash); token != -1; token = map.nextToken(hash, token)) {
-            PoolEntry e = map.getElementByToken(token);
+            PoolEntry e = entryByIndex(map.getIndexByToken(token));
             if (e.tag() == TAG_DOUBLE
                     && e instanceof AbstractPoolEntry.DoubleEntryImpl ce
                     && ce.doubleValue() == val)
@@ -337,9 +328,9 @@ private DoubleEntry findDoubleEntry(double val) {
     private<T extends AbstractPoolEntry> AbstractPoolEntry findEntry(int tag, T ref1) {
         // invariant: canWriteDirect(ref1.constantPool())
         int hash = AbstractPoolEntry.hash1(tag, ref1.index());
-        EntryMap<PoolEntry> map = map();
+        EntryMap map = map();
         for (int token = map.firstToken(hash); token != -1; token = map.nextToken(hash, token)) {
-            PoolEntry e = map.getElementByToken(token);
+            PoolEntry e = entryByIndex(map.getIndexByToken(token));
             if (e.tag() == tag
                 && e instanceof AbstractPoolEntry.AbstractRefEntry<?> re
                 && re.ref1 == ref1)
@@ -356,9 +347,9 @@ private<T extends AbstractPoolEntry> AbstractPoolEntry findEntry(int tag, T ref1
             AbstractPoolEntry findEntry(int tag, T ref1, U ref2) {
         // invariant: canWriteDirect(ref1.constantPool()), canWriteDirect(ref2.constantPool())
         int hash = AbstractPoolEntry.hash2(tag, ref1.index(), ref2.index());
-        EntryMap<PoolEntry> map = map();
+        EntryMap map = map();
         for (int token = map.firstToken(hash); token != -1; token = map.nextToken(hash, token)) {
-            PoolEntry e = map.getElementByToken(token);
+            PoolEntry e = entryByIndex(map.getIndexByToken(token));
             if (e.tag() == tag
                     && e instanceof AbstractPoolEntry.AbstractRefsEntry<?, ?> re
                     && re.ref1 == ref1
@@ -374,10 +365,10 @@ AbstractPoolEntry findEntry(int tag, T ref1, U ref2) {
     }
 
     private AbstractPoolEntry.Utf8EntryImpl tryFindUtf8(int hash, String target) {
-        EntryMap<PoolEntry> map = map();
+        EntryMap map = map();
         for (int token = map.firstToken(hash); token != -1;
              token = map.nextToken(hash, token)) {
-            PoolEntry e = map.getElementByToken(token);
+            PoolEntry e = entryByIndex(map.getIndexByToken(token));
             if (e.tag() == ClassFile.TAG_UTF8
                 && e instanceof AbstractPoolEntry.Utf8EntryImpl ce
                 && ce.hashCode() == hash
@@ -392,9 +383,9 @@ private AbstractPoolEntry.Utf8EntryImpl tryFindUtf8(int hash, String target) {
     }
 
     private AbstractPoolEntry.Utf8EntryImpl tryFindUtf8(int hash, AbstractPoolEntry.Utf8EntryImpl target) {
-        EntryMap<PoolEntry> map = map();
+        EntryMap map = map();
         for (int token = map.firstToken(hash); token != -1; token = map.nextToken(hash, token)) {
-            PoolEntry e = map.getElementByToken(token);
+            PoolEntry e = entryByIndex(map.getIndexByToken(token));
             if (e.tag() == ClassFile.TAG_UTF8
                 && e instanceof AbstractPoolEntry.Utf8EntryImpl ce
                 && target.equalsUtf8(ce))
@@ -407,6 +398,20 @@ private AbstractPoolEntry.Utf8EntryImpl tryFindUtf8(int hash, AbstractPoolEntry.
         return null;
     }
 
+    @Override
+    public Utf8Entry utf8Entry(ClassDesc desc) {
+        var utf8 = utf8Entry(desc.descriptorString());
+        utf8.typeSym = desc;
+        return utf8;
+    }
+
+    @Override
+    public Utf8Entry utf8Entry(MethodTypeDesc desc) {
+        var utf8 = utf8Entry(desc.descriptorString());
+        utf8.typeSym = desc;
+        return utf8;
+    }
+
     @Override
     public AbstractPoolEntry.Utf8EntryImpl utf8Entry(String s) {
         int hash = AbstractPoolEntry.hashString(s.hashCode());
@@ -489,9 +494,7 @@ public InterfaceMethodRefEntry interfaceMethodRefEntry(ClassEntry owner, NameAnd
 
     @Override
     public MethodTypeEntry methodTypeEntry(MethodTypeDesc descriptor) {
-        var ret = (AbstractPoolEntry.MethodTypeEntryImpl)methodTypeEntry(utf8Entry(descriptor.descriptorString()));
-        ret.sym = descriptor;
-        return ret;
+        return methodTypeEntry(utf8Entry(descriptor));
     }
 
     @Override
@@ -513,9 +516,9 @@ public MethodHandleEntry methodHandleEntry(int refKind, MemberRefEntry reference
         }
 
         int hash = AbstractPoolEntry.hash2(TAG_METHODHANDLE, refKind, reference.index());
-        EntryMap<PoolEntry> map1 = map();
+        EntryMap map1 = map();
         for (int token = map1.firstToken(hash); token != -1; token = map1.nextToken(hash, token)) {
-            PoolEntry e = map1.getElementByToken(token);
+            PoolEntry e = entryByIndex(map1.getIndexByToken(token));
             if (e.tag() == TAG_METHODHANDLE
                 && e instanceof AbstractPoolEntry.MethodHandleEntryImpl ce
                 && ce.kind() == refKind && ce.reference() == reference)
@@ -539,9 +542,9 @@ public InvokeDynamicEntry invokeDynamicEntry(BootstrapMethodEntry bootstrapMetho
             nameAndType = nameAndTypeEntry(nameAndType.name(), nameAndType.type());
         int hash = AbstractPoolEntry.hash2(TAG_INVOKEDYNAMIC,
                 bootstrapMethodEntry.bsmIndex(), nameAndType.index());
-        EntryMap<PoolEntry> map1 = map();
+        EntryMap map1 = map();
         for (int token = map1.firstToken(hash); token != -1; token = map1.nextToken(hash, token)) {
-            PoolEntry e = map1.getElementByToken(token);
+            PoolEntry e = entryByIndex(map1.getIndexByToken(token));
             if (e.tag() == TAG_INVOKEDYNAMIC
                 && e instanceof AbstractPoolEntry.InvokeDynamicEntryImpl ce
                 && ce.bootstrap() == bootstrapMethodEntry && ce.nameAndType() == nameAndType)
@@ -570,9 +573,9 @@ public ConstantDynamicEntry constantDynamicEntry(BootstrapMethodEntry bootstrapM
             nameAndType = nameAndTypeEntry(nameAndType.name(), nameAndType.type());
         int hash = AbstractPoolEntry.hash2(TAG_CONSTANTDYNAMIC,
                 bootstrapMethodEntry.bsmIndex(), nameAndType.index());
-        EntryMap<PoolEntry> map1 = map();
+        EntryMap map1 = map();
         for (int token = map1.firstToken(hash); token != -1; token = map1.nextToken(hash, token)) {
-            PoolEntry e = map1.getElementByToken(token);
+            PoolEntry e = entryByIndex(map1.getIndexByToken(token));
             if (e.tag() == TAG_CONSTANTDYNAMIC
                 && e instanceof AbstractPoolEntry.ConstantDynamicEntryImpl ce
                 && ce.bootstrap() == bootstrapMethodEntry && ce.nameAndType() == nameAndType)
@@ -640,9 +643,9 @@ public BootstrapMethodEntry bsmEntry(MethodHandleEntry methodReference,
         }
         AbstractPoolEntry.MethodHandleEntryImpl mre = (AbstractPoolEntry.MethodHandleEntryImpl) methodReference;
         int hash = BootstrapMethodEntryImpl.computeHashCode(mre, arguments);
-        EntryMap<BootstrapMethodEntryImpl> map = bsmMap();
+        EntryMap map = bsmMap();
         for (int token = map.firstToken(hash); token != -1; token = map.nextToken(hash, token)) {
-            BootstrapMethodEntryImpl e = map.getElementByToken(token);
+            BootstrapMethodEntryImpl e = bootstrapMethodEntry(map.getIndexByToken(token));
             if (e.bootstrapMethod() == mre && e.arguments().equals(arguments)) {
                 return e;
             }
diff --git a/src/java.base/share/classes/jdk/internal/classfile/impl/Util.java b/src/java.base/share/classes/jdk/internal/classfile/impl/Util.java
index b0f1de65d9291..7f3dd914e7f8c 100644
--- a/src/java.base/share/classes/jdk/internal/classfile/impl/Util.java
+++ b/src/java.base/share/classes/jdk/internal/classfile/impl/Util.java
@@ -30,6 +30,7 @@
 import java.lang.classfile.MethodBuilder;
 import java.lang.classfile.PseudoInstruction;
 import java.lang.classfile.constantpool.PoolEntry;
+import java.lang.classfile.constantpool.Utf8Entry;
 import java.lang.constant.ClassDesc;
 import java.lang.constant.MethodTypeDesc;
 import java.util.AbstractList;
@@ -220,12 +221,20 @@ public static boolean has(AccessFlag.Location location, int flagsMask, AccessFla
         return (flag.mask() & flagsMask) == flag.mask() && flag.locations().contains(location);
     }
 
+    public static ClassDesc fieldTypeSymbol(Utf8Entry utf8) {
+        return ((AbstractPoolEntry.Utf8EntryImpl) utf8).fieldTypeSymbol();
+    }
+
+    public static MethodTypeDesc methodTypeSymbol(Utf8Entry utf8) {
+        return ((AbstractPoolEntry.Utf8EntryImpl) utf8).methodTypeSymbol();
+    }
+
     public static ClassDesc fieldTypeSymbol(NameAndTypeEntry nat) {
-        return ((AbstractPoolEntry.NameAndTypeEntryImpl)nat).fieldTypeSymbol();
+        return fieldTypeSymbol(nat.type());
     }
 
     public static MethodTypeDesc methodTypeSymbol(NameAndTypeEntry nat) {
-        return ((AbstractPoolEntry.NameAndTypeEntryImpl)nat).methodTypeSymbol();
+        return methodTypeSymbol(nat.type());
     }
 
     @SuppressWarnings("unchecked")
diff --git a/src/java.base/share/classes/jdk/internal/foreign/AbstractMemorySegmentImpl.java b/src/java.base/share/classes/jdk/internal/foreign/AbstractMemorySegmentImpl.java
index 83b11b7ce686b..64994af5cb790 100644
--- a/src/java.base/share/classes/jdk/internal/foreign/AbstractMemorySegmentImpl.java
+++ b/src/java.base/share/classes/jdk/internal/foreign/AbstractMemorySegmentImpl.java
@@ -72,8 +72,6 @@ public abstract sealed class AbstractMemorySegmentImpl
         implements MemorySegment, SegmentAllocator, BiFunction<String, List<Number>, RuntimeException>
         permits HeapMemorySegmentImpl, NativeMemorySegmentImpl {
 
-    private static final ScopedMemoryAccess SCOPED_MEMORY_ACCESS = ScopedMemoryAccess.getScopedMemoryAccess();
-
     static final JavaNioAccess NIO_ACCESS = SharedSecrets.getJavaNioAccess();
 
     final long length;
@@ -189,53 +187,10 @@ public Stream<MemorySegment> elements(MemoryLayout elementLayout) {
         return StreamSupport.stream(spliterator(elementLayout), false);
     }
 
-    // FILL_NATIVE_THRESHOLD must be a power of two and should be greater than 2^3
-    // Update the value for Aarch64 once 8338975 is fixed.
-    private static final long FILL_NATIVE_THRESHOLD = 1L << (Architecture.isAARCH64() ? 10 : 5);
-
-    @Override
     @ForceInline
+    @Override
     public final MemorySegment fill(byte value) {
-        checkReadOnly(false);
-        if (length == 0) {
-            // Implicit state check
-            checkValidState();
-        } else if (length < FILL_NATIVE_THRESHOLD) {
-            // 0 <= length < FILL_NATIVE_LIMIT : 0...0X...XXXX
-
-            // Handle smaller segments directly without transitioning to native code
-            final long u = Byte.toUnsignedLong(value);
-            final long longValue = u << 56 | u << 48 | u << 40 | u << 32 | u << 24 | u << 16 | u << 8 | u;
-
-            int offset = 0;
-            // 0...0X...X000
-            final int limit = (int) (length & (FILL_NATIVE_THRESHOLD - 8));
-            for (; offset < limit; offset += 8) {
-                SCOPED_MEMORY_ACCESS.putLong(sessionImpl(), unsafeGetBase(), unsafeGetOffset() + offset, longValue);
-            }
-            int remaining = (int) length - limit;
-            // 0...0X00
-            if (remaining >= 4) {
-                SCOPED_MEMORY_ACCESS.putInt(sessionImpl(), unsafeGetBase(), unsafeGetOffset() + offset, (int) longValue);
-                offset += 4;
-                remaining -= 4;
-            }
-            // 0...00X0
-            if (remaining >= 2) {
-                SCOPED_MEMORY_ACCESS.putShort(sessionImpl(), unsafeGetBase(), unsafeGetOffset() + offset, (short) longValue);
-                offset += 2;
-                remaining -= 2;
-            }
-            // 0...000X
-            if (remaining == 1) {
-                SCOPED_MEMORY_ACCESS.putByte(sessionImpl(), unsafeGetBase(), unsafeGetOffset() + offset, value);
-            }
-            // We have now fully handled 0...0X...XXXX
-        } else {
-            // Handle larger segments via native calls
-            SCOPED_MEMORY_ACCESS.setMemory(sessionImpl(), unsafeGetBase(), unsafeGetOffset(), length, value);
-        }
-        return this;
+        return SegmentBulkOperations.fill(this, value);
     }
 
     @Override
@@ -244,38 +199,6 @@ public MemorySegment allocate(long byteSize, long byteAlignment) {
         return asSlice(0, byteSize, byteAlignment);
     }
 
-    /**
-     * Mismatch over long lengths.
-     */
-    public static long vectorizedMismatchLargeForBytes(MemorySessionImpl aSession, MemorySessionImpl bSession,
-                                                        Object a, long aOffset,
-                                                        Object b, long bOffset,
-                                                        long length) {
-        long off = 0;
-        long remaining = length;
-        int i, size;
-        boolean lastSubRange = false;
-        while (remaining > 7 && !lastSubRange) {
-            if (remaining > Integer.MAX_VALUE) {
-                size = Integer.MAX_VALUE;
-            } else {
-                size = (int) remaining;
-                lastSubRange = true;
-            }
-            i = SCOPED_MEMORY_ACCESS.vectorizedMismatch(aSession, bSession,
-                    a, aOffset + off,
-                    b, bOffset + off,
-                    size, ArraysSupport.LOG2_ARRAY_BYTE_INDEX_SCALE);
-            if (i >= 0)
-                return off + i;
-
-            i = size - ~i;
-            off += i;
-            remaining -= i;
-        }
-        return ~remaining;
-    }
-
     @Override
     public final ByteBuffer asByteBuffer() {
         checkArraySize("ByteBuffer", 1);
@@ -314,7 +237,7 @@ public final Optional<MemorySegment> asOverlappingSlice(MemorySegment other) {
     }
 
     @ForceInline
-    private boolean overlaps(AbstractMemorySegmentImpl that) {
+    boolean overlaps(AbstractMemorySegmentImpl that) {
         if (unsafeGetBase() == that.unsafeGetBase()) {  // both either native or the same heap segment
             final long thisStart = this.unsafeGetOffset();
             final long thatStart = that.unsafeGetOffset();
@@ -334,7 +257,8 @@ public MemorySegment copyFrom(MemorySegment src) {
     @Override
     public long mismatch(MemorySegment other) {
         Objects.requireNonNull(other);
-        return MemorySegment.mismatch(this, 0, byteSize(), other, 0, other.byteSize());
+        return SegmentBulkOperations.mismatch(this, 0, byteSize(),
+                (AbstractMemorySegmentImpl) other, 0, other.byteSize());
     }
 
     @Override
@@ -650,64 +574,6 @@ private static Object bufferRef(Buffer buffer) {
         }
     }
 
-    // COPY_NATIVE_THRESHOLD must be a power of two and should be greater than 2^3
-    private static final long COPY_NATIVE_THRESHOLD = 1 << 6;
-
-    @ForceInline
-    public static void copy(AbstractMemorySegmentImpl src, long srcOffset,
-                            AbstractMemorySegmentImpl dst, long dstOffset,
-                            long size) {
-
-        Utils.checkNonNegativeIndex(size, "size");
-        // Implicit null check for src and dst
-        src.checkAccess(srcOffset, size, true);
-        dst.checkAccess(dstOffset, size, false);
-
-        if (size <= 0) {
-            // Do nothing
-        } else if (size < COPY_NATIVE_THRESHOLD && !src.overlaps(dst)) {
-            // 0 < size < FILL_NATIVE_LIMIT : 0...0X...XXXX
-            //
-            // Strictly, we could check for !src.asSlice(srcOffset, size).overlaps(dst.asSlice(dstOffset, size) but
-            // this is a bit slower and it likely very unusual there is any difference in the outcome. Also, if there
-            // is an overlap, we could tolerate one particular direction of overlap (but not the other).
-
-            // 0...0X...X000
-            final int limit = (int) (size & (COPY_NATIVE_THRESHOLD - 8));
-            int offset = 0;
-            for (; offset < limit; offset += 8) {
-                final long v = SCOPED_MEMORY_ACCESS.getLong(src.sessionImpl(), src.unsafeGetBase(), src.unsafeGetOffset() + srcOffset + offset);
-                SCOPED_MEMORY_ACCESS.putLong(dst.sessionImpl(), dst.unsafeGetBase(), dst.unsafeGetOffset() + dstOffset + offset, v);
-            }
-            int remaining = (int) size - offset;
-            // 0...0X00
-            if (remaining >= 4) {
-                final int v = SCOPED_MEMORY_ACCESS.getInt(src.sessionImpl(), src.unsafeGetBase(),src.unsafeGetOffset() + srcOffset + offset);
-                SCOPED_MEMORY_ACCESS.putInt(dst.sessionImpl(), dst.unsafeGetBase(), dst.unsafeGetOffset() + dstOffset + offset, v);
-                offset += 4;
-                remaining -= 4;
-            }
-            // 0...00X0
-            if (remaining >= 2) {
-                final short v = SCOPED_MEMORY_ACCESS.getShort(src.sessionImpl(), src.unsafeGetBase(), src.unsafeGetOffset() + srcOffset + offset);
-                SCOPED_MEMORY_ACCESS.putShort(dst.sessionImpl(), dst.unsafeGetBase(), dst.unsafeGetOffset() + dstOffset + offset, v);
-                offset += 2;
-                remaining -=2;
-            }
-            // 0...000X
-            if (remaining == 1) {
-                final byte v = SCOPED_MEMORY_ACCESS.getByte(src.sessionImpl(), src.unsafeGetBase(), src.unsafeGetOffset() + srcOffset + offset);
-                SCOPED_MEMORY_ACCESS.putByte(dst.sessionImpl(), dst.unsafeGetBase(), dst.unsafeGetOffset() + dstOffset + offset, v);
-            }
-            // We have now fully handled 0...0X...XXXX
-        } else {
-            // For larger sizes, the transition to native code pays off
-            SCOPED_MEMORY_ACCESS.copyMemory(src.sessionImpl(), dst.sessionImpl(),
-                    src.unsafeGetBase(), src.unsafeGetOffset() + srcOffset,
-                    dst.unsafeGetBase(), dst.unsafeGetOffset() + dstOffset, size);
-        }
-    }
-
     @ForceInline
     public static void copy(MemorySegment srcSegment, ValueLayout srcElementLayout, long srcOffset,
                             MemorySegment dstSegment, ValueLayout dstElementLayout, long dstOffset,
@@ -794,40 +660,6 @@ public static void copy(Object srcArray, int srcIndex,
         }
     }
 
-    public static long mismatch(MemorySegment srcSegment, long srcFromOffset, long srcToOffset,
-                                MemorySegment dstSegment, long dstFromOffset, long dstToOffset) {
-        AbstractMemorySegmentImpl srcImpl = (AbstractMemorySegmentImpl)Objects.requireNonNull(srcSegment);
-        AbstractMemorySegmentImpl dstImpl = (AbstractMemorySegmentImpl)Objects.requireNonNull(dstSegment);
-        long srcBytes = srcToOffset - srcFromOffset;
-        long dstBytes = dstToOffset - dstFromOffset;
-        srcImpl.checkAccess(srcFromOffset, srcBytes, true);
-        dstImpl.checkAccess(dstFromOffset, dstBytes, true);
-
-        long bytes = Math.min(srcBytes, dstBytes);
-        long i = 0;
-        if (bytes > 7) {
-            if (srcImpl.get(JAVA_BYTE, srcFromOffset) != dstImpl.get(JAVA_BYTE, dstFromOffset)) {
-                return 0;
-            }
-            i = AbstractMemorySegmentImpl.vectorizedMismatchLargeForBytes(srcImpl.sessionImpl(), dstImpl.sessionImpl(),
-                    srcImpl.unsafeGetBase(), srcImpl.unsafeGetOffset() + srcFromOffset,
-                    dstImpl.unsafeGetBase(), dstImpl.unsafeGetOffset() + dstFromOffset,
-                    bytes);
-            if (i >= 0) {
-                return i;
-            }
-            long remaining = ~i;
-            assert remaining < 8 : "remaining greater than 7: " + remaining;
-            i = bytes - remaining;
-        }
-        for (; i < bytes; i++) {
-            if (srcImpl.get(JAVA_BYTE, srcFromOffset + i) != dstImpl.get(JAVA_BYTE, dstFromOffset + i)) {
-                return i;
-            }
-        }
-        return srcBytes != dstBytes ? bytes : -1;
-    }
-
     private static int getScaleFactor(Buffer buffer) {
         return switch (buffer) {
             case ByteBuffer   _                 -> 0;
diff --git a/src/java.base/share/classes/jdk/internal/foreign/SegmentBulkOperations.java b/src/java.base/share/classes/jdk/internal/foreign/SegmentBulkOperations.java
new file mode 100644
index 0000000000000..41ae247293d3f
--- /dev/null
+++ b/src/java.base/share/classes/jdk/internal/foreign/SegmentBulkOperations.java
@@ -0,0 +1,317 @@
+/*
+ * Copyright (c) 2024, Oracle and/or its affiliates. All rights reserved.
+ * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
+ *
+ * This code is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 only, as
+ * published by the Free Software Foundation.  Oracle designates this
+ * particular file as subject to the "Classpath" exception as provided
+ * by Oracle in the LICENSE file that accompanied this code.
+ *
+ * This code is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+ * version 2 for more details (a copy is included in the LICENSE file that
+ * accompanied this code).
+ *
+ * You should have received a copy of the GNU General Public License version
+ * 2 along with this work; if not, write to the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
+ * or visit www.oracle.com if you need additional information or have any
+ * questions.
+ */
+
+package jdk.internal.foreign;
+
+import jdk.internal.misc.ScopedMemoryAccess;
+import jdk.internal.util.Architecture;
+import jdk.internal.util.ArraysSupport;
+import jdk.internal.vm.annotation.ForceInline;
+import sun.security.action.GetIntegerAction;
+
+import java.lang.foreign.MemorySegment;
+
+/**
+ * This class contains optimized bulk operation methods that operate on one or several
+ * memory segments.
+ * <p>
+ * Generally, the methods attempt to work with as-large-as-possible units of memory at
+ * a time.
+ * <p>
+ * It should be noted that when invoking scoped memory access get/set operations, it
+ * is imperative from a performance perspective to convey the sharp types from the
+ * call site in order for the compiler to pick the correct Unsafe access variant.
+ */
+public final class SegmentBulkOperations {
+
+    private SegmentBulkOperations() {}
+
+    private static final ScopedMemoryAccess SCOPED_MEMORY_ACCESS = ScopedMemoryAccess.getScopedMemoryAccess();
+
+    // All the threshold values below MUST be a power of two and should preferably be
+    // greater or equal to 2^3.
+
+    // Update the FILL value for Aarch64 once 8338975 is fixed.
+    private static final int NATIVE_THRESHOLD_FILL = powerOfPropertyOr("fill", Architecture.isAARCH64() ? 10 : 5);
+    private static final int NATIVE_THRESHOLD_MISMATCH = powerOfPropertyOr("mismatch", 6);
+    private static final int NATIVE_THRESHOLD_COPY = powerOfPropertyOr("copy", 6);
+
+    @ForceInline
+    public static MemorySegment fill(AbstractMemorySegmentImpl dst, byte value) {
+        dst.checkReadOnly(false);
+        if (dst.length == 0) {
+            // Implicit state check
+            dst.checkValidState();
+        } else if (dst.length < NATIVE_THRESHOLD_FILL) {
+            // 0 <= length < FILL_NATIVE_LIMIT : 0...0X...XXXX
+
+            // Handle smaller segments directly without transitioning to native code
+            final long u = Byte.toUnsignedLong(value);
+            final long longValue = u << 56 | u << 48 | u << 40 | u << 32 | u << 24 | u << 16 | u << 8 | u;
+
+            int offset = 0;
+            // 0...0X...X000
+            final int limit = (int) (dst.length & (NATIVE_THRESHOLD_FILL - 8));
+            for (; offset < limit; offset += 8) {
+                SCOPED_MEMORY_ACCESS.putLongUnaligned(dst.sessionImpl(), dst.unsafeGetBase(), dst.unsafeGetOffset() + offset, longValue, !Architecture.isLittleEndian());
+            }
+            int remaining = (int) dst.length - limit;
+            // 0...0X00
+            if (remaining >= 4) {
+                SCOPED_MEMORY_ACCESS.putIntUnaligned(dst.sessionImpl(), dst.unsafeGetBase(), dst.unsafeGetOffset() + offset, (int) longValue, !Architecture.isLittleEndian());
+                offset += 4;
+                remaining -= 4;
+            }
+            // 0...00X0
+            if (remaining >= 2) {
+                SCOPED_MEMORY_ACCESS.putShortUnaligned(dst.sessionImpl(), dst.unsafeGetBase(), dst.unsafeGetOffset() + offset, (short) longValue, !Architecture.isLittleEndian());
+                offset += 2;
+                remaining -= 2;
+            }
+            // 0...000X
+            if (remaining == 1) {
+                SCOPED_MEMORY_ACCESS.putByte(dst.sessionImpl(), dst.unsafeGetBase(), dst.unsafeGetOffset() + offset, value);
+            }
+            // We have now fully handled 0...0X...XXXX
+        } else {
+            // Handle larger segments via native calls
+            SCOPED_MEMORY_ACCESS.setMemory(dst.sessionImpl(), dst.unsafeGetBase(), dst.unsafeGetOffset(), dst.length, value);
+        }
+        return dst;
+    }
+
+    @ForceInline
+    public static void copy(AbstractMemorySegmentImpl src, long srcOffset,
+                            AbstractMemorySegmentImpl dst, long dstOffset,
+                            long size) {
+
+        Utils.checkNonNegativeIndex(size, "size");
+        // Implicit null check for src and dst
+        src.checkAccess(srcOffset, size, true);
+        dst.checkAccess(dstOffset, size, false);
+
+        if (size <= 0) {
+            // Do nothing
+        } else if (size < NATIVE_THRESHOLD_COPY && !src.overlaps(dst)) {
+            // 0 < size < FILL_NATIVE_LIMIT : 0...0X...XXXX
+            //
+            // Strictly, we could check for !src.asSlice(srcOffset, size).overlaps(dst.asSlice(dstOffset, size) but
+            // this is a bit slower and it likely very unusual there is any difference in the outcome. Also, if there
+            // is an overlap, we could tolerate one particular direction of overlap (but not the other).
+
+            // 0...0X...X000
+            final int limit = (int) (size & (NATIVE_THRESHOLD_COPY - 8));
+            int offset = 0;
+            for (; offset < limit; offset += 8) {
+                final long v = SCOPED_MEMORY_ACCESS.getLongUnaligned(src.sessionImpl(), src.unsafeGetBase(), src.unsafeGetOffset() + srcOffset + offset, !Architecture.isLittleEndian());
+                SCOPED_MEMORY_ACCESS.putLongUnaligned(dst.sessionImpl(), dst.unsafeGetBase(), dst.unsafeGetOffset() + dstOffset + offset, v, !Architecture.isLittleEndian());
+            }
+            int remaining = (int) size - offset;
+            // 0...0X00
+            if (remaining >= 4) {
+                final int v = SCOPED_MEMORY_ACCESS.getIntUnaligned(src.sessionImpl(), src.unsafeGetBase(),src.unsafeGetOffset() + srcOffset + offset, !Architecture.isLittleEndian());
+                SCOPED_MEMORY_ACCESS.putIntUnaligned(dst.sessionImpl(), dst.unsafeGetBase(), dst.unsafeGetOffset() + dstOffset + offset, v, !Architecture.isLittleEndian());
+                offset += 4;
+                remaining -= 4;
+            }
+            // 0...00X0
+            if (remaining >= 2) {
+                final short v = SCOPED_MEMORY_ACCESS.getShortUnaligned(src.sessionImpl(), src.unsafeGetBase(), src.unsafeGetOffset() + srcOffset + offset, !Architecture.isLittleEndian());
+                SCOPED_MEMORY_ACCESS.putShortUnaligned(dst.sessionImpl(), dst.unsafeGetBase(), dst.unsafeGetOffset() + dstOffset + offset, v, !Architecture.isLittleEndian());
+                offset += 2;
+                remaining -=2;
+            }
+            // 0...000X
+            if (remaining == 1) {
+                final byte v = SCOPED_MEMORY_ACCESS.getByte(src.sessionImpl(), src.unsafeGetBase(), src.unsafeGetOffset() + srcOffset + offset);
+                SCOPED_MEMORY_ACCESS.putByte(dst.sessionImpl(), dst.unsafeGetBase(), dst.unsafeGetOffset() + dstOffset + offset, v);
+            }
+            // We have now fully handled 0...0X...XXXX
+        } else {
+            // For larger sizes, the transition to native code pays off
+            SCOPED_MEMORY_ACCESS.copyMemory(src.sessionImpl(), dst.sessionImpl(),
+                    src.unsafeGetBase(), src.unsafeGetOffset() + srcOffset,
+                    dst.unsafeGetBase(), dst.unsafeGetOffset() + dstOffset, size);
+        }
+    }
+
+    @ForceInline
+    public static long mismatch(AbstractMemorySegmentImpl src, long srcFromOffset, long srcToOffset,
+                                AbstractMemorySegmentImpl dst, long dstFromOffset, long dstToOffset) {
+        final long srcBytes = srcToOffset - srcFromOffset;
+        final long dstBytes = dstToOffset - dstFromOffset;
+        src.checkAccess(srcFromOffset, srcBytes, true);
+        dst.checkAccess(dstFromOffset, dstBytes, true);
+
+        final long length = Math.min(srcBytes, dstBytes);
+        final boolean srcAndDstBytesDiffer = srcBytes != dstBytes;
+
+        if (length == 0) {
+            return srcAndDstBytesDiffer ? 0 : -1;
+        } else if (length < NATIVE_THRESHOLD_MISMATCH) {
+            return mismatch(src, srcFromOffset, dst, dstFromOffset, 0, (int) length, srcAndDstBytesDiffer);
+        } else {
+            long i;
+            if (SCOPED_MEMORY_ACCESS.getByte(src.sessionImpl(), src.unsafeGetBase(), src.unsafeGetOffset() + srcFromOffset) !=
+                    SCOPED_MEMORY_ACCESS.getByte(dst.sessionImpl(), dst.unsafeGetBase(), dst.unsafeGetOffset() + dstFromOffset)) {
+                return 0;
+            }
+            i = vectorizedMismatchLargeForBytes(src.sessionImpl(), dst.sessionImpl(),
+                    src.unsafeGetBase(), src.unsafeGetOffset() + srcFromOffset,
+                    dst.unsafeGetBase(), dst.unsafeGetOffset() + dstFromOffset,
+                    length);
+            if (i >= 0) {
+                return i;
+            }
+            final long remaining = ~i;
+            assert remaining < 8 : "remaining greater than 7: " + remaining;
+            i = length - remaining;
+            return mismatch(src, srcFromOffset + i, dst, dstFromOffset + i, i, (int) remaining, srcAndDstBytesDiffer);
+        }
+    }
+
+    // Mismatch is handled in chunks of 64 (unroll of eight 8s), 8, 4, 2, and 1 byte(s).
+    @ForceInline
+    private static long mismatch(AbstractMemorySegmentImpl src, long srcFromOffset,
+                                 AbstractMemorySegmentImpl dst, long dstFromOffset,
+                                 long start, int length, boolean srcAndDstBytesDiffer) {
+        int offset = 0;
+        final int limit = length & (NATIVE_THRESHOLD_MISMATCH - 8);
+        for (; offset < limit; offset += 8) {
+            final long s = SCOPED_MEMORY_ACCESS.getLongUnaligned(src.sessionImpl(), src.unsafeGetBase(), src.unsafeGetOffset() + srcFromOffset + offset, !Architecture.isLittleEndian());
+            final long d = SCOPED_MEMORY_ACCESS.getLongUnaligned(dst.sessionImpl(), dst.unsafeGetBase(), dst.unsafeGetOffset() + dstFromOffset + offset, !Architecture.isLittleEndian());
+            if (s != d) {
+                return start + offset + mismatch(s, d);
+            }
+        }
+        int remaining = length - offset;
+        // 0...XXX000
+        for (; remaining >= 8; remaining -= 8) {
+            final long s = SCOPED_MEMORY_ACCESS.getLongUnaligned(src.sessionImpl(), src.unsafeGetBase(), src.unsafeGetOffset() + srcFromOffset + offset, !Architecture.isLittleEndian());
+            final long d = SCOPED_MEMORY_ACCESS.getLongUnaligned(dst.sessionImpl(), dst.unsafeGetBase(), dst.unsafeGetOffset() + dstFromOffset + offset, !Architecture.isLittleEndian());
+            if (s != d) {
+                return start + offset + mismatch(s, d);
+            }
+            offset += 8;
+        }
+
+        // 0...0X00
+        if (remaining >= 4) {
+            final int s = SCOPED_MEMORY_ACCESS.getIntUnaligned(src.sessionImpl(), src.unsafeGetBase(), src.unsafeGetOffset() + srcFromOffset + offset, !Architecture.isLittleEndian());
+            final int d = SCOPED_MEMORY_ACCESS.getIntUnaligned(dst.sessionImpl(), dst.unsafeGetBase(), dst.unsafeGetOffset() + dstFromOffset + offset, !Architecture.isLittleEndian());
+            if (s != d) {
+                return start + offset + mismatch(s, d);
+            }
+            offset += 4;
+            remaining -= 4;
+        }
+        // 0...00X0
+        if (remaining >= 2) {
+            final short s = SCOPED_MEMORY_ACCESS.getShortUnaligned(src.sessionImpl(), src.unsafeGetBase(), src.unsafeGetOffset() + srcFromOffset + offset, !Architecture.isLittleEndian());
+            final short d = SCOPED_MEMORY_ACCESS.getShortUnaligned(dst.sessionImpl(), dst.unsafeGetBase(), dst.unsafeGetOffset() + dstFromOffset + offset, !Architecture.isLittleEndian());
+            if (s != d) {
+                return start + offset + mismatch(s, d);
+            }
+            offset += 2;
+            remaining -= 2;
+        }
+        // 0...000X
+        if (remaining == 1) {
+            final byte s = SCOPED_MEMORY_ACCESS.getByte(src.sessionImpl(), src.unsafeGetBase(), src.unsafeGetOffset() + srcFromOffset + offset);
+            final byte d = SCOPED_MEMORY_ACCESS.getByte(dst.sessionImpl(), dst.unsafeGetBase(), dst.unsafeGetOffset() + dstFromOffset + offset);
+            if (s != d) {
+                return start + offset;
+            }
+        }
+        return srcAndDstBytesDiffer ? (start + length) : -1;
+        // We have now fully handled 0...0X...XXXX
+    }
+
+    @ForceInline
+    private static int mismatch(long first, long second) {
+        final long x = first ^ second;
+        return (Architecture.isLittleEndian()
+                ? Long.numberOfTrailingZeros(x)
+                : Long.numberOfLeadingZeros(x)) / 8;
+    }
+
+    @ForceInline
+    private static int mismatch(int first, int second) {
+        final int x = first ^ second;
+        return (Architecture.isLittleEndian()
+                ? Integer.numberOfTrailingZeros(x)
+                : Integer.numberOfLeadingZeros(x)) / 8;
+    }
+
+    @ForceInline
+    private static int mismatch(short first, short second) {
+        if (Architecture.isLittleEndian()) {
+            return ((0xff & first) == (0xff & second)) ? 1 : 0;
+        } else {
+            return ((0xff & first) == (0xff & second)) ? 0 : 1;
+        }
+    }
+
+    /**
+     * Mismatch over long lengths.
+     */
+    private static long vectorizedMismatchLargeForBytes(MemorySessionImpl aSession, MemorySessionImpl bSession,
+                                                        Object a, long aOffset,
+                                                        Object b, long bOffset,
+                                                        long length) {
+        long off = 0;
+        long remaining = length;
+        int i, size;
+        boolean lastSubRange = false;
+        while (remaining > 7 && !lastSubRange) {
+            if (remaining > Integer.MAX_VALUE) {
+                size = Integer.MAX_VALUE;
+            } else {
+                size = (int) remaining;
+                lastSubRange = true;
+            }
+            i = SCOPED_MEMORY_ACCESS.vectorizedMismatch(aSession, bSession,
+                    a, aOffset + off,
+                    b, bOffset + off,
+                    size, ArraysSupport.LOG2_ARRAY_BYTE_INDEX_SCALE);
+            if (i >= 0)
+                return off + i;
+
+            i = size - ~i;
+            off += i;
+            remaining -= i;
+        }
+        return ~remaining;
+    }
+
+    static final String PROPERTY_PATH = "java.lang.foreign.native.threshold.power.";
+
+    // The returned value is in the interval [0, 2^30]
+    static int powerOfPropertyOr(String name, int defaultPower) {
+        final int power = GetIntegerAction.privilegedGetProperty(PROPERTY_PATH + name, defaultPower);
+        return 1 << Math.clamp(power, 0, Integer.SIZE - 2);
+    }
+
+}
diff --git a/src/jdk.compiler/share/classes/com/sun/tools/javac/code/Flags.java b/src/jdk.compiler/share/classes/com/sun/tools/javac/code/Flags.java
index 346aab18a60fd..a2937b73a143f 100644
--- a/src/jdk.compiler/share/classes/com/sun/tools/javac/code/Flags.java
+++ b/src/jdk.compiler/share/classes/com/sun/tools/javac/code/Flags.java
@@ -285,7 +285,12 @@ public static EnumSet<Flag> asFlagSet(long flags) {
     /**
      * Flag that marks a synthetic method body for a lambda expression
      */
-    public static final long LAMBDA_METHOD = 1L<<49;
+    public static final long LAMBDA_METHOD = 1L<<49; //MethodSymbols only
+
+    /**
+     * Flag that marks a synthetic local capture field in a local/anon class
+     */
+    public static final long LOCAL_CAPTURE_FIELD = 1L<<49; //VarSymbols only
 
     /**
      * Flag to control recursion in TransTypes
diff --git a/src/jdk.compiler/share/classes/com/sun/tools/javac/comp/Attr.java b/src/jdk.compiler/share/classes/com/sun/tools/javac/comp/Attr.java
index 44ccbade4a900..07a136c17003c 100644
--- a/src/jdk.compiler/share/classes/com/sun/tools/javac/comp/Attr.java
+++ b/src/jdk.compiler/share/classes/com/sun/tools/javac/comp/Attr.java
@@ -182,6 +182,7 @@ protected Attr(Context context) {
         unknownTypeInfo = new ResultInfo(KindSelector.TYP, Type.noType);
         unknownTypeExprInfo = new ResultInfo(KindSelector.VAL_TYP, Type.noType);
         recoveryInfo = new RecoveryInfo(deferredAttr.emptyDeferredAttrContext);
+        initBlockType = new MethodType(List.nil(), syms.voidType, List.nil(), syms.methodClass);
     }
 
     /** Switch: reifiable types in instanceof enabled?
@@ -628,6 +629,7 @@ public void report(DiagnosticPosition pos, JCDiagnostic details) {
     final ResultInfo unknownTypeInfo;
     final ResultInfo unknownTypeExprInfo;
     final ResultInfo recoveryInfo;
+    final MethodType initBlockType;
 
     Type pt() {
         return resultInfo.pt;
@@ -1421,7 +1423,7 @@ public void visitBlock(JCBlock tree) {
             // created BLOCK-method.
             Symbol fakeOwner =
                 new MethodSymbol(tree.flags | BLOCK |
-                    env.info.scope.owner.flags() & STRICTFP, names.empty, null,
+                    env.info.scope.owner.flags() & STRICTFP, names.empty, initBlockType,
                     env.info.scope.owner);
             final Env<AttrContext> localEnv =
                 env.dup(tree, env.info.dup(env.info.scope.dupUnshared(fakeOwner)));
@@ -3524,62 +3526,26 @@ void checkLambdaCompatible(JCLambda tree, Type descriptor, CheckContext checkCon
             }
         }
 
-        /* Map to hold 'fake' clinit methods. If a lambda is used to initialize a
-         * static field and that lambda has type annotations, these annotations will
-         * also be stored at these fake clinit methods.
-         *
-         * LambdaToMethod also use fake clinit methods so they can be reused.
-         * Also as LTM is a phase subsequent to attribution, the methods from
-         * clinits can be safely removed by LTM to save memory.
-         */
-        private Map<ClassSymbol, MethodSymbol> clinits = new HashMap<>();
-
-        public MethodSymbol removeClinit(ClassSymbol sym) {
-            return clinits.remove(sym);
-        }
-
         /* This method returns an environment to be used to attribute a lambda
          * expression.
          *
          * The owner of this environment is a method symbol. If the current owner
-         * is not a method, for example if the lambda is used to initialize
-         * a field, then if the field is:
-         *
-         * - an instance field, we use the first constructor.
-         * - a static field, we create a fake clinit method.
+         * is not a method (e.g. if the lambda occurs in a field initializer), then
+         * a synthetic method symbol owner is created.
          */
         public Env<AttrContext> lambdaEnv(JCLambda that, Env<AttrContext> env) {
             Env<AttrContext> lambdaEnv;
             Symbol owner = env.info.scope.owner;
             if (owner.kind == VAR && owner.owner.kind == TYP) {
-                //field initializer
+                // If the lambda is nested in a field initializer, we need to create a fake init method.
+                // Uniqueness of this symbol is not important (as e.g. annotations will be added on the
+                // init symbol's owner).
                 ClassSymbol enclClass = owner.enclClass();
-                Symbol newScopeOwner = env.info.scope.owner;
-                /* if the field isn't static, then we can get the first constructor
-                 * and use it as the owner of the environment. This is what
-                 * LTM code is doing to look for type annotations so we are fine.
-                 */
-                if ((owner.flags() & STATIC) == 0) {
-                    for (Symbol s : enclClass.members_field.getSymbolsByName(names.init)) {
-                        newScopeOwner = s;
-                        break;
-                    }
-                } else {
-                    /* if the field is static then we need to create a fake clinit
-                     * method, this method can later be reused by LTM.
-                     */
-                    MethodSymbol clinit = clinits.get(enclClass);
-                    if (clinit == null) {
-                        Type clinitType = new MethodType(List.nil(),
-                                syms.voidType, List.nil(), syms.methodClass);
-                        clinit = new MethodSymbol(STATIC | SYNTHETIC | PRIVATE,
-                                names.clinit, clinitType, enclClass);
-                        clinit.params = List.nil();
-                        clinits.put(enclClass, clinit);
-                    }
-                    newScopeOwner = clinit;
-                }
-                lambdaEnv = env.dup(that, env.info.dup(env.info.scope.dupUnshared(newScopeOwner)));
+                Name initName = owner.isStatic() ? names.clinit : names.init;
+                MethodSymbol initSym = new MethodSymbol(BLOCK | (owner.isStatic() ? STATIC : 0) | SYNTHETIC | PRIVATE,
+                        initName, initBlockType, enclClass);
+                initSym.params = List.nil();
+                lambdaEnv = env.dup(that, env.info.dup(env.info.scope.dupUnshared(initSym)));
             } else {
                 lambdaEnv = env.dup(that, env.info.dup(env.info.scope.dup()));
             }
@@ -3936,6 +3902,7 @@ private void setFunctionalInfo(final Env<AttrContext> env, final JCFunctionalExp
                     inferenceContext -> setFunctionalInfo(env, fExpr, pt, inferenceContext.asInstType(descriptorType),
                     inferenceContext.asInstType(primaryTarget), checkContext));
         } else {
+            fExpr.owner = env.info.scope.owner;
             if (pt.hasTag(CLASS)) {
                 fExpr.target = primaryTarget;
             }
diff --git a/src/jdk.compiler/share/classes/com/sun/tools/javac/comp/CaptureScanner.java b/src/jdk.compiler/share/classes/com/sun/tools/javac/comp/CaptureScanner.java
new file mode 100644
index 0000000000000..ef8a3ff584de6
--- /dev/null
+++ b/src/jdk.compiler/share/classes/com/sun/tools/javac/comp/CaptureScanner.java
@@ -0,0 +1,99 @@
+/*
+ * Copyright (c) 2024, Oracle and/or its affiliates. All rights reserved.
+ * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
+ *
+ * This code is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 only, as
+ * published by the Free Software Foundation.  Oracle designates this
+ * particular file as subject to the "Classpath" exception as provided
+ * by Oracle in the LICENSE file that accompanied this code.
+ *
+ * This code is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+ * version 2 for more details (a copy is included in the LICENSE file that
+ * accompanied this code).
+ *
+ * You should have received a copy of the GNU General Public License version
+ * 2 along with this work; if not, write to the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
+ * or visit www.oracle.com if you need additional information or have any
+ * questions.
+ */
+
+package com.sun.tools.javac.comp;
+
+import com.sun.tools.javac.code.Symbol;
+import com.sun.tools.javac.code.Symbol.VarSymbol;
+import com.sun.tools.javac.tree.JCTree;
+import com.sun.tools.javac.tree.TreeScanner;
+import com.sun.tools.javac.util.List;
+
+import java.util.HashSet;
+import java.util.LinkedHashSet;
+import java.util.SequencedSet;
+import java.util.Set;
+
+import static com.sun.tools.javac.code.Kinds.Kind.MTH;
+import static com.sun.tools.javac.code.Kinds.Kind.VAR;
+
+/**
+ * A visitor which collects the set of local variables "captured" by a given tree.
+ */
+public class CaptureScanner extends TreeScanner {
+
+    /**
+     * The tree under analysis.
+     */
+    private final JCTree tree;
+
+    /**
+     * The set of local variable declarations encountered in the tree under analysis.
+     */
+    private final Set<Symbol.VarSymbol> seenVars = new HashSet<>();
+
+    /**
+     * The set of captured local variables accessed from within the tree under analysis.
+     */
+    private final SequencedSet<VarSymbol> fvs = new LinkedHashSet<>();
+
+    public CaptureScanner(JCTree ownerTree) {
+        this.tree = ownerTree;
+    }
+
+    @Override
+    public void visitIdent(JCTree.JCIdent tree) {
+        Symbol sym = tree.sym;
+        if (sym.kind == VAR && sym.owner.kind == MTH) {
+            Symbol.VarSymbol vsym = (Symbol.VarSymbol) sym;
+            if (vsym.getConstValue() == null && !seenVars.contains(vsym)) {
+                addFreeVar(vsym);
+            }
+        }
+    }
+
+    /**
+     * Add free variable to fvs list unless it is already there.
+     */
+    protected void addFreeVar(Symbol.VarSymbol v) {
+        fvs.add(v);
+    }
+
+    @Override
+    public void visitVarDef(JCTree.JCVariableDecl tree) {
+        if (tree.sym.owner.kind == MTH) {
+            seenVars.add(tree.sym);
+        }
+        super.visitVarDef(tree);
+    }
+
+    /**
+     * Obtains the list of captured local variables in the tree under analysis.
+     */
+    List<Symbol.VarSymbol> analyzeCaptures() {
+        scan(tree);
+        return List.from(fvs);
+    }
+}
diff --git a/src/jdk.compiler/share/classes/com/sun/tools/javac/comp/LambdaToMethod.java b/src/jdk.compiler/share/classes/com/sun/tools/javac/comp/LambdaToMethod.java
index 2c3d79c0ab6da..adfc3ceaa0d94 100644
--- a/src/jdk.compiler/share/classes/com/sun/tools/javac/comp/LambdaToMethod.java
+++ b/src/jdk.compiler/share/classes/com/sun/tools/javac/comp/LambdaToMethod.java
@@ -25,50 +25,87 @@
 
 package com.sun.tools.javac.comp;
 
-import com.sun.tools.javac.code.Symbol.MethodHandleSymbol;
-import com.sun.tools.javac.code.Types.SignatureGenerator.InvalidSignatureException;
-import com.sun.tools.javac.jvm.PoolConstant.LoadableConstant;
-import com.sun.tools.javac.resources.CompilerProperties.Errors;
-import com.sun.tools.javac.resources.CompilerProperties.Fragments;
-import com.sun.tools.javac.tree.*;
-import com.sun.tools.javac.tree.JCTree.*;
-import com.sun.tools.javac.tree.JCTree.JCMemberReference.ReferenceKind;
-import com.sun.tools.javac.tree.TreeMaker;
-import com.sun.tools.javac.tree.TreeTranslator;
 import com.sun.tools.javac.code.Attribute;
+import com.sun.tools.javac.code.Flags;
 import com.sun.tools.javac.code.Symbol;
 import com.sun.tools.javac.code.Symbol.ClassSymbol;
 import com.sun.tools.javac.code.Symbol.DynamicMethodSymbol;
+import com.sun.tools.javac.code.Symbol.MethodHandleSymbol;
 import com.sun.tools.javac.code.Symbol.MethodSymbol;
 import com.sun.tools.javac.code.Symbol.VarSymbol;
 import com.sun.tools.javac.code.Symtab;
 import com.sun.tools.javac.code.Type;
 import com.sun.tools.javac.code.Type.MethodType;
 import com.sun.tools.javac.code.Types;
-import com.sun.tools.javac.comp.LambdaToMethod.LambdaAnalyzerPreprocessor.*;
+import com.sun.tools.javac.code.Types.SignatureGenerator.InvalidSignatureException;
+import com.sun.tools.javac.jvm.PoolConstant.LoadableConstant;
+import com.sun.tools.javac.main.Option;
+import com.sun.tools.javac.resources.CompilerProperties.Errors;
+import com.sun.tools.javac.resources.CompilerProperties.Fragments;
 import com.sun.tools.javac.resources.CompilerProperties.Notes;
-import com.sun.tools.javac.util.*;
+import com.sun.tools.javac.tree.JCTree;
+import com.sun.tools.javac.tree.JCTree.JCAnnotation;
+import com.sun.tools.javac.tree.JCTree.JCBinary;
+import com.sun.tools.javac.tree.JCTree.JCBlock;
+import com.sun.tools.javac.tree.JCTree.JCBreak;
+import com.sun.tools.javac.tree.JCTree.JCCase;
+import com.sun.tools.javac.tree.JCTree.JCClassDecl;
+import com.sun.tools.javac.tree.JCTree.JCExpression;
+import com.sun.tools.javac.tree.JCTree.JCFieldAccess;
+import com.sun.tools.javac.tree.JCTree.JCFunctionalExpression;
+import com.sun.tools.javac.tree.JCTree.JCIdent;
+import com.sun.tools.javac.tree.JCTree.JCLambda;
+import com.sun.tools.javac.tree.JCTree.JCMemberReference;
+import com.sun.tools.javac.tree.JCTree.JCMethodDecl;
+import com.sun.tools.javac.tree.JCTree.JCMethodInvocation;
+import com.sun.tools.javac.tree.JCTree.JCNewClass;
+import com.sun.tools.javac.tree.JCTree.JCReturn;
+import com.sun.tools.javac.tree.JCTree.JCStatement;
+import com.sun.tools.javac.tree.JCTree.JCSwitch;
+import com.sun.tools.javac.tree.JCTree.JCVariableDecl;
+import com.sun.tools.javac.tree.JCTree.Tag;
+import com.sun.tools.javac.tree.TreeInfo;
+import com.sun.tools.javac.tree.TreeMaker;
+import com.sun.tools.javac.tree.TreeTranslator;
+import com.sun.tools.javac.util.Assert;
+import com.sun.tools.javac.util.Context;
+import com.sun.tools.javac.util.DiagnosticSource;
+import com.sun.tools.javac.util.InvalidUtfException;
+import com.sun.tools.javac.util.JCDiagnostic;
 import com.sun.tools.javac.util.JCDiagnostic.DiagnosticPosition;
+import com.sun.tools.javac.util.List;
+import com.sun.tools.javac.util.ListBuffer;
+import com.sun.tools.javac.util.Log;
+import com.sun.tools.javac.util.Name;
+import com.sun.tools.javac.util.Names;
+import com.sun.tools.javac.util.Options;
 
-import java.util.EnumMap;
+import javax.lang.model.element.ElementKind;
+import java.lang.invoke.LambdaMetafactory;
 import java.util.HashMap;
 import java.util.HashSet;
-import java.util.LinkedHashMap;
 import java.util.Map;
-import java.util.Optional;
 import java.util.Set;
 import java.util.function.Consumer;
 import java.util.function.Supplier;
 
-import static com.sun.tools.javac.comp.LambdaToMethod.LambdaSymbolKind.*;
-import static com.sun.tools.javac.code.Flags.*;
-import static com.sun.tools.javac.code.Kinds.Kind.*;
-import static com.sun.tools.javac.code.TypeTag.*;
-import static com.sun.tools.javac.tree.JCTree.Tag.*;
-
-import javax.lang.model.element.ElementKind;
-
-import com.sun.tools.javac.main.Option;
+import static com.sun.tools.javac.code.Flags.ABSTRACT;
+import static com.sun.tools.javac.code.Flags.BLOCK;
+import static com.sun.tools.javac.code.Flags.DEFAULT;
+import static com.sun.tools.javac.code.Flags.FINAL;
+import static com.sun.tools.javac.code.Flags.INTERFACE;
+import static com.sun.tools.javac.code.Flags.LAMBDA_METHOD;
+import static com.sun.tools.javac.code.Flags.LOCAL_CAPTURE_FIELD;
+import static com.sun.tools.javac.code.Flags.PARAMETER;
+import static com.sun.tools.javac.code.Flags.PRIVATE;
+import static com.sun.tools.javac.code.Flags.STATIC;
+import static com.sun.tools.javac.code.Flags.STRICTFP;
+import static com.sun.tools.javac.code.Flags.SYNTHETIC;
+import static com.sun.tools.javac.code.Kinds.Kind.MTH;
+import static com.sun.tools.javac.code.Kinds.Kind.TYP;
+import static com.sun.tools.javac.code.Kinds.Kind.VAR;
+import static com.sun.tools.javac.code.TypeTag.BOT;
+import static com.sun.tools.javac.code.TypeTag.VOID;
 
 /**
  * This pass desugars lambda expressions into static methods
@@ -80,31 +117,28 @@
  */
 public class LambdaToMethod extends TreeTranslator {
 
-    private Attr attr;
-    private JCDiagnostic.Factory diags;
-    private Log log;
-    private Lower lower;
-    private Names names;
-    private Symtab syms;
-    private Resolve rs;
-    private Operators operators;
+    private final Attr attr;
+    private final JCDiagnostic.Factory diags;
+    private final Log log;
+    private final Lower lower;
+    private final Names names;
+    private final Symtab syms;
+    private final Resolve rs;
+    private final Operators operators;
     private TreeMaker make;
-    private Types types;
-    private TransTypes transTypes;
+    private final Types types;
+    private final TransTypes transTypes;
     private Env<AttrContext> attrEnv;
 
-    /** the analyzer scanner */
-    private LambdaAnalyzerPreprocessor analyzer;
-
-    /** map from lambda trees to translation contexts */
-    private Map<JCTree, TranslationContext<?>> contextMap;
-
-    /** current translation context (visitor argument) */
-    private TranslationContext<?> context;
-
     /** info about the current class being processed */
     private KlassInfo kInfo;
 
+    /** translation context of the current lambda expression */
+    private LambdaTranslationContext lambdaContext;
+
+    /** the variable whose initializer is pending */
+    private VarSymbol pendingVar;
+
     /** dump statistics about lambda code generation */
     private final boolean dumpLambdaToMethodStats;
 
@@ -121,13 +155,13 @@ public class LambdaToMethod extends TreeTranslator {
     private final boolean deduplicateLambdas;
 
     /** Flag for alternate metafactories indicating the lambda object is intended to be serializable */
-    public static final int FLAG_SERIALIZABLE = 1 << 0;
+    public static final int FLAG_SERIALIZABLE = LambdaMetafactory.FLAG_SERIALIZABLE;
 
     /** Flag for alternate metafactories indicating the lambda object has multiple targets */
-    public static final int FLAG_MARKERS = 1 << 1;
+    public static final int FLAG_MARKERS = LambdaMetafactory.FLAG_MARKERS;
 
     /** Flag for alternate metafactories indicating the lambda object requires multiple bridges */
-    public static final int FLAG_BRIDGES = 1 << 2;
+    public static final int FLAG_BRIDGES = LambdaMetafactory.FLAG_BRIDGES;
 
     // <editor-fold defaultstate="collapsed" desc="Instantiating">
     protected static final Context.Key<LambdaToMethod> unlambdaKey = new Context.Key<>();
@@ -151,18 +185,17 @@ private LambdaToMethod(Context context) {
         make = TreeMaker.instance(context);
         types = Types.instance(context);
         transTypes = TransTypes.instance(context);
-        analyzer = new LambdaAnalyzerPreprocessor();
         Options options = Options.instance(context);
         dumpLambdaToMethodStats = options.isSet("debug.dumpLambdaToMethodStats");
         attr = Attr.instance(context);
         forceSerializable = options.isSet("forceSerializable");
         boolean lineDebugInfo =
-            options.isUnset(Option.G_CUSTOM) ||
-            options.isSet(Option.G_CUSTOM, "lines");
+                options.isUnset(Option.G_CUSTOM) ||
+                        options.isSet(Option.G_CUSTOM, "lines");
         boolean varDebugInfo =
-            options.isUnset(Option.G_CUSTOM)
-            ? options.isSet(Option.G)
-            : options.isSet(Option.G_CUSTOM, "vars");
+                options.isUnset(Option.G_CUSTOM)
+                        ? options.isSet(Option.G)
+                        : options.isSet(Option.G_CUSTOM, "vars");
         debugLinesOrVars = lineDebugInfo || varDebugInfo;
         verboseDeduplication = options.isSet("debug.dumpLambdaToMethodDeduplication");
         deduplicateLambdas = options.getBoolean("deduplicateLambdas", true);
@@ -180,7 +213,6 @@ class DedupedLambda {
             this.tree = tree;
         }
 
-
         @Override
         public int hashCode() {
             int hashCode = this.hashCode;
@@ -203,18 +235,18 @@ private class KlassInfo {
         /**
          * list of methods to append
          */
-        private ListBuffer<JCTree> appendedMethodList;
+        private ListBuffer<JCTree> appendedMethodList = new ListBuffer<>();
 
-        private Map<DedupedLambda, DedupedLambda> dedupedLambdas;
+        private final Map<DedupedLambda, DedupedLambda> dedupedLambdas = new HashMap<>();
 
-        private Map<Object, DynamicMethodSymbol> dynMethSyms = new HashMap<>();
+        private final Map<Object, DynamicMethodSymbol> dynMethSyms = new HashMap<>();
 
         /**
          * list of deserialization cases
          */
-        private final Map<String, ListBuffer<JCStatement>> deserializeCases;
+        private final Map<String, ListBuffer<JCStatement>> deserializeCases = new HashMap<>();
 
-       /**
+        /**
          * deserialize method symbol
          */
         private final MethodSymbol deserMethodSym;
@@ -226,11 +258,10 @@ private class KlassInfo {
 
         private final JCClassDecl clazz;
 
+        private final Map<String, Integer> syntheticNames = new HashMap<>();
+
         private KlassInfo(JCClassDecl clazz) {
             this.clazz = clazz;
-            appendedMethodList = new ListBuffer<>();
-            dedupedLambdas = new HashMap<>();
-            deserializeCases = new HashMap<>();
             MethodType type = new MethodType(List.of(syms.serializedLambdaType), syms.objectType,
                     List.nil(), syms.methodClass);
             deserMethodSym = makePrivateSyntheticMethod(STATIC, names.deserializeLambda, type, clazz.sym);
@@ -241,62 +272,50 @@ private KlassInfo(JCClassDecl clazz) {
         private void addMethod(JCTree decl) {
             appendedMethodList = appendedMethodList.prepend(decl);
         }
-    }
-
-    // <editor-fold defaultstate="collapsed" desc="translate methods">
-    @Override
-    public <T extends JCTree> T translate(T tree) {
-        TranslationContext<?> newContext = contextMap.get(tree);
-        return translate(tree, newContext != null ? newContext : context);
-    }
 
-    <T extends JCTree> T translate(T tree, TranslationContext<?> newContext) {
-        TranslationContext<?> prevContext = context;
-        try {
-            context = newContext;
-            return super.translate(tree);
-        }
-        finally {
-            context = prevContext;
-        }
-    }
-
-    <T extends JCTree> List<T> translate(List<T> trees, TranslationContext<?> newContext) {
-        ListBuffer<T> buf = new ListBuffer<>();
-        for (T tree : trees) {
-            buf.append(translate(tree, newContext));
+        int syntheticNameIndex(StringBuilder buf, int start) {
+            String temp = buf.toString();
+            Integer count = syntheticNames.get(temp);
+            if (count == null) {
+                count = start;
+            }
+            syntheticNames.put(temp, count + 1);
+            return count;
         }
-        return buf.toList();
     }
 
+    // <editor-fold defaultstate="collapsed" desc="visitor methods">
     public JCTree translateTopLevelClass(Env<AttrContext> env, JCTree cdef, TreeMaker make) {
         this.make = make;
         this.attrEnv = env;
-        this.context = null;
-        this.contextMap = new HashMap<>();
-        cdef = analyzer.analyzeAndPreprocessClass((JCClassDecl) cdef);
         return translate(cdef);
     }
-    // </editor-fold>
 
-    // <editor-fold defaultstate="collapsed" desc="visitor methods">
     /**
      * Visit a class.
      * Maintain the translatedMethodList across nested classes.
      * Append the translatedMethodList to the class after it is translated.
-     * @param tree
      */
     @Override
     public void visitClassDef(JCClassDecl tree) {
         KlassInfo prevKlassInfo = kInfo;
+        DiagnosticSource prevSource = log.currentSource();
+        LambdaTranslationContext prevLambdaContext = lambdaContext;
+        VarSymbol prevPendingVar = pendingVar;
         try {
             kInfo = new KlassInfo(tree);
+            log.useSource(tree.sym.sourcefile);
+            lambdaContext = null;
+            pendingVar = null;
             super.visitClassDef(tree);
+            if (prevLambdaContext != null) {
+                tree.sym.owner = prevLambdaContext.translatedSym;
+            }
             if (!kInfo.deserializeCases.isEmpty()) {
                 int prevPos = make.pos;
                 try {
                     make.at(tree);
-                    kInfo.addMethod(makeDeserializeMethod(tree.sym));
+                    kInfo.addMethod(makeDeserializeMethod());
                 } finally {
                     make.at(prevPos);
                 }
@@ -310,6 +329,9 @@ public void visitClassDef(JCClassDecl tree) {
             result = tree;
         } finally {
             kInfo = prevKlassInfo;
+            log.useSource(prevSource.getFile());
+            lambdaContext = prevLambdaContext;
+            pendingVar = prevPendingVar;
         }
     }
 
@@ -317,11 +339,10 @@ public void visitClassDef(JCClassDecl tree) {
      * Translate a lambda into a method to be inserted into the class.
      * Then replace the lambda site with an invokedynamic call of to lambda
      * meta-factory, which will use the lambda method.
-     * @param tree
      */
     @Override
     public void visitLambda(JCLambda tree) {
-        LambdaTranslationContext localContext = (LambdaTranslationContext)context;
+        LambdaTranslationContext localContext = new LambdaTranslationContext(tree);
         MethodSymbol sym = localContext.translatedSym;
         MethodType lambdaType = (MethodType) sym.type;
 
@@ -332,26 +353,26 @@ public void visitLambda(JCLambda tree) {
                lambda and attach it to the implementation method.
             */
 
-            Symbol owner = localContext.owner;
+            Symbol owner = tree.owner;
             apportionTypeAnnotations(tree,
                     owner::getRawTypeAttributes,
                     owner::setTypeAttributes,
                     sym::setTypeAttributes);
 
-
-            boolean init;
-            if ((init = (owner.name == names.init)) || owner.name == names.clinit) {
-                owner = owner.owner;
+            final long ownerFlags = owner.flags();
+            if ((ownerFlags & Flags.BLOCK) != 0) {
+                ClassSymbol cs = (ClassSymbol) owner.owner;
+                boolean isStaticInit = (ownerFlags & Flags.STATIC) != 0;
                 apportionTypeAnnotations(tree,
-                        init ? owner::getInitTypeAttributes : owner::getClassInitTypeAttributes,
-                        init ? owner::setInitTypeAttributes : owner::setClassInitTypeAttributes,
+                        isStaticInit ? cs::getClassInitTypeAttributes : cs::getInitTypeAttributes,
+                        isStaticInit ? cs::setClassInitTypeAttributes : cs::setInitTypeAttributes,
                         sym::appendUniqueTypeAttributes);
             }
-            if (localContext.self != null && localContext.self.getKind() == ElementKind.FIELD) {
-                owner = localContext.self;
+
+            if (pendingVar != null && pendingVar.getKind() == ElementKind.FIELD) {
                 apportionTypeAnnotations(tree,
-                        owner::getRawTypeAttributes,
-                        owner::setTypeAttributes,
+                        pendingVar::getRawTypeAttributes,
+                        pendingVar::setTypeAttributes,
                         sym::appendUniqueTypeAttributes);
             }
         }
@@ -363,35 +384,13 @@ public void visitLambda(JCLambda tree) {
                 List.nil(),
                 localContext.syntheticParams,
                 lambdaType.getThrownTypes() == null ?
-                    List.nil() :
-                    make.Types(lambdaType.getThrownTypes()),
+                        List.nil() :
+                        make.Types(lambdaType.getThrownTypes()),
                 null,
                 null);
         lambdaDecl.sym = sym;
         lambdaDecl.type = lambdaType;
 
-        //translate lambda body
-        //As the lambda body is translated, all references to lambda locals,
-        //captured variables, enclosing members are adjusted accordingly
-        //to refer to the static method parameters (rather than i.e. accessing
-        //captured members directly).
-        lambdaDecl.body = translate(makeLambdaBody(tree, lambdaDecl));
-
-        boolean dedupe = false;
-        if (deduplicateLambdas && !debugLinesOrVars && !localContext.isSerializable()) {
-            DedupedLambda dedupedLambda = new DedupedLambda(lambdaDecl.sym, lambdaDecl.body);
-            DedupedLambda existing = kInfo.dedupedLambdas.putIfAbsent(dedupedLambda, dedupedLambda);
-            if (existing != null) {
-                sym = existing.symbol;
-                dedupe = true;
-                if (verboseDeduplication) log.note(tree, Notes.VerboseL2mDeduplicate(sym));
-            }
-        }
-        if (!dedupe) {
-            //Add the method to the list of methods to be added to this class.
-            kInfo.addMethod(lambdaDecl);
-        }
-
         //now that we have generated a method for the lambda expression,
         //we can translate the lambda into a method reference pointing to the newly
         //created method.
@@ -408,46 +407,72 @@ public void visitLambda(JCLambda tree) {
         if (!sym.isStatic()) {
             syntheticInits.append(makeThis(
                     sym.owner.enclClass().asType(),
-                    localContext.owner.enclClass()));
+                    tree.owner.enclClass()));
         }
 
         //add captured locals
-        for (Symbol fv : localContext.getSymbolMap(CAPTURED_VAR).keySet()) {
-            if (fv != localContext.self) {
-                JCExpression captured_local = make.Ident(fv).setType(fv.type);
-                syntheticInits.append(captured_local);
-            }
+        for (Symbol fv : localContext.capturedVars) {
+            JCExpression captured_local = make.Ident(fv).setType(fv.type);
+            syntheticInits.append(captured_local);
         }
 
         //then, determine the arguments to the indy call
-        List<JCExpression> indy_args = translate(syntheticInits.toList(), localContext.prev);
+        List<JCExpression> indy_args = translate(syntheticInits.toList());
+
+        LambdaTranslationContext prevLambdaContext = lambdaContext;
+        try {
+            lambdaContext = localContext;
+            //translate lambda body
+            //As the lambda body is translated, all references to lambda locals,
+            //captured variables, enclosing members are adjusted accordingly
+            //to refer to the static method parameters (rather than i.e. accessing
+            //captured members directly).
+            lambdaDecl.body = translate(makeLambdaBody(tree, lambdaDecl));
+        } finally {
+            lambdaContext = prevLambdaContext;
+        }
+
+        boolean dedupe = false;
+        if (deduplicateLambdas && !debugLinesOrVars && !isSerializable(tree)) {
+            DedupedLambda dedupedLambda = new DedupedLambda(lambdaDecl.sym, lambdaDecl.body);
+            DedupedLambda existing = kInfo.dedupedLambdas.putIfAbsent(dedupedLambda, dedupedLambda);
+            if (existing != null) {
+                sym = existing.symbol;
+                dedupe = true;
+                if (verboseDeduplication) log.note(tree, Notes.VerboseL2mDeduplicate(sym));
+            }
+        }
+        if (!dedupe) {
+            //Add the method to the list of methods to be added to this class.
+            kInfo.addMethod(lambdaDecl);
+        }
 
         //convert to an invokedynamic call
-        result = makeMetafactoryIndyCall(context, sym.asHandle(), indy_args);
+        result = makeMetafactoryIndyCall(tree, sym.asHandle(), localContext.translatedSym, indy_args);
     }
 
     // where
-        // Reassign type annotations from the source that should really belong to the lambda
-        private void apportionTypeAnnotations(JCLambda tree,
-                                              Supplier<List<Attribute.TypeCompound>> source,
-                                              Consumer<List<Attribute.TypeCompound>> owner,
-                                              Consumer<List<Attribute.TypeCompound>> lambda) {
-
-            ListBuffer<Attribute.TypeCompound> ownerTypeAnnos = new ListBuffer<>();
-            ListBuffer<Attribute.TypeCompound> lambdaTypeAnnos = new ListBuffer<>();
-
-            for (Attribute.TypeCompound tc : source.get()) {
-                if (tc.position.onLambda == tree) {
-                    lambdaTypeAnnos.append(tc);
-                } else {
-                    ownerTypeAnnos.append(tc);
-                }
-            }
-            if (lambdaTypeAnnos.nonEmpty()) {
-                owner.accept(ownerTypeAnnos.toList());
-                lambda.accept(lambdaTypeAnnos.toList());
+    // Reassign type annotations from the source that should really belong to the lambda
+    private void apportionTypeAnnotations(JCLambda tree,
+                                          Supplier<List<Attribute.TypeCompound>> source,
+                                          Consumer<List<Attribute.TypeCompound>> owner,
+                                          Consumer<List<Attribute.TypeCompound>> lambda) {
+
+        ListBuffer<Attribute.TypeCompound> ownerTypeAnnos = new ListBuffer<>();
+        ListBuffer<Attribute.TypeCompound> lambdaTypeAnnos = new ListBuffer<>();
+
+        for (Attribute.TypeCompound tc : source.get()) {
+            if (tc.position.onLambda == tree) {
+                lambdaTypeAnnos.append(tc);
+            } else {
+                ownerTypeAnnos.append(tc);
             }
         }
+        if (lambdaTypeAnnos.nonEmpty()) {
+            owner.accept(ownerTypeAnnos.toList());
+            lambda.accept(lambdaTypeAnnos.toList());
+        }
+    }
 
     private JCIdent makeThis(Type type, Symbol owner) {
         VarSymbol _this = new VarSymbol(PARAMETER | FINAL | SYNTHETIC,
@@ -460,65 +485,46 @@ private JCIdent makeThis(Type type, Symbol owner) {
     /**
      * Translate a method reference into an invokedynamic call to the
      * meta-factory.
-     * @param tree
      */
     @Override
     public void visitReference(JCMemberReference tree) {
-        ReferenceTranslationContext localContext = (ReferenceTranslationContext)context;
-
         //first determine the method symbol to be used to generate the sam instance
         //this is either the method reference symbol, or the bridged reference symbol
         MethodSymbol refSym = (MethodSymbol)tree.sym;
 
         //the qualifying expression is treated as a special captured arg
-        JCExpression init;
-        switch(tree.kind) {
-
-            case IMPLICIT_INNER:    /** Inner :: new */
-            case SUPER:             /** super :: instMethod */
-                init = makeThis(
-                    localContext.owner.enclClass().asType(),
-                    localContext.owner.enclClass());
-                break;
-
-            case BOUND:             /** Expr :: instMethod */
-                init = transTypes.coerce(attrEnv, tree.getQualifierExpression(),
-                    types.erasure(tree.sym.owner.type));
-                init = attr.makeNullCheck(init);
-                break;
-
-            case UNBOUND:           /** Type :: instMethod */
-            case STATIC:            /** Type :: staticMethod */
-            case TOPLEVEL:          /** Top level :: new */
-            case ARRAY_CTOR:        /** ArrayType :: new */
-                init = null;
-                break;
-
-            default:
-                throw new InternalError("Should not have an invalid kind");
-        }
-
-        List<JCExpression> indy_args = init==null? List.nil() : translate(List.of(init), localContext.prev);
-
+        JCExpression init = switch (tree.kind) {
+            case IMPLICIT_INNER,    /* Inner :: new */
+                 SUPER ->           /* super :: instMethod */
+                    makeThis(tree.owner.enclClass().asType(), tree.owner.enclClass());
+            case BOUND ->           /* Expr :: instMethod */
+                    attr.makeNullCheck(transTypes.coerce(attrEnv, tree.getQualifierExpression(),
+                            types.erasure(tree.sym.owner.type)));
+            case UNBOUND,           /* Type :: instMethod */
+                 STATIC,            /* Type :: staticMethod */
+                 TOPLEVEL,          /* Top level :: new */
+                 ARRAY_CTOR ->      /* ArrayType :: new */
+                    null;
+        };
+
+        List<JCExpression> indy_args = (init == null) ?
+                List.nil() : translate(List.of(init));
 
         //build a sam instance using an indy call to the meta-factory
-        result = makeMetafactoryIndyCall(localContext, refSym.asHandle(), indy_args);
+        result = makeMetafactoryIndyCall(tree, refSym.asHandle(), refSym, indy_args);
     }
 
     /**
      * Translate identifiers within a lambda to the mapped identifier
-     * @param tree
      */
     @Override
     public void visitIdent(JCIdent tree) {
-        if (context == null || !analyzer.lambdaIdentSymbolFilter(tree.sym)) {
+        if (lambdaContext == null) {
             super.visitIdent(tree);
         } else {
             int prevPos = make.pos;
             try {
                 make.at(tree);
-
-                LambdaTranslationContext lambdaContext = (LambdaTranslationContext) context;
                 JCTree ltree = lambdaContext.translate(tree);
                 if (ltree != null) {
                     result = ltree;
@@ -535,13 +541,18 @@ public void visitIdent(JCIdent tree) {
 
     @Override
     public void visitVarDef(JCVariableDecl tree) {
-        LambdaTranslationContext lambdaContext = (LambdaTranslationContext)context;
-        if (context != null && lambdaContext.getSymbolMap(LOCAL_VAR).containsKey(tree.sym)) {
-            tree.init = translate(tree.init);
-            tree.sym = (VarSymbol) lambdaContext.getSymbolMap(LOCAL_VAR).get(tree.sym);
-            result = tree;
-        } else {
-            super.visitVarDef(tree);
+        VarSymbol prevPendingVar = pendingVar;
+        try {
+            pendingVar = tree.sym;
+            if (lambdaContext != null) {
+                tree.sym = lambdaContext.addLocal(tree.sym);
+                tree.init = translate(tree.init);
+                result = tree;
+            } else {
+                super.visitVarDef(tree);
+            }
+        } finally {
+            pendingVar = prevPendingVar;
         }
     }
 
@@ -609,7 +620,7 @@ public void visitReturn(JCReturn tree) {
                 if (isTarget_void && !isLambda_void) {
                     //Void to void conversion:
                     // { TYPE $loc = RET-EXPR; return; }
-                    VarSymbol loc = makeSyntheticVar(0, names.fromString("$loc"), tree.expr.type, lambdaMethodDecl.sym);
+                    VarSymbol loc = new VarSymbol(SYNTHETIC, names.fromString("$loc"), tree.expr.type, lambdaMethodDecl.sym);
                     JCVariableDecl varDef = make.VarDef(loc, tree.expr);
                     result = make.Block(0, List.of(varDef, make.Return(null)));
                 } else {
@@ -628,7 +639,7 @@ public void visitReturn(JCReturn tree) {
         return trans_block;
     }
 
-    private JCMethodDecl makeDeserializeMethod(Symbol kSym) {
+    private JCMethodDecl makeDeserializeMethod() {
         ListBuffer<JCCase> cases = new ListBuffer<>();
         ListBuffer<JCBreak> breaks = new ListBuffer<>();
         for (Map.Entry<String, ListBuffer<JCStatement>> entry : kInfo.deserializeCases.entrySet()) {
@@ -644,16 +655,16 @@ private JCMethodDecl makeDeserializeMethod(Symbol kSym) {
         JCBlock body = make.Block(0L, List.of(
                 sw,
                 make.Throw(makeNewClass(
-                    syms.illegalArgumentExceptionType,
-                    List.of(make.Literal("Invalid lambda deserialization"))))));
+                        syms.illegalArgumentExceptionType,
+                        List.of(make.Literal("Invalid lambda deserialization"))))));
         JCMethodDecl deser = make.MethodDef(make.Modifiers(kInfo.deserMethodSym.flags()),
-                        names.deserializeLambda,
-                        make.QualIdent(kInfo.deserMethodSym.getReturnType().tsym),
-                        List.nil(),
-                        List.of(make.VarDef(kInfo.deserParamSym, null)),
-                        List.nil(),
-                        body,
-                        null);
+                names.deserializeLambda,
+                make.QualIdent(kInfo.deserMethodSym.getReturnType().tsym),
+                List.nil(),
+                List.of(make.VarDef(kInfo.deserParamSym, null)),
+                List.nil(),
+                body,
+                null);
         deser.sym = kInfo.deserMethodSym;
         deser.type = kInfo.deserMethodSym.type;
         //System.err.printf("DESER: '%s'\n", deser);
@@ -667,7 +678,7 @@ private JCMethodDecl makeDeserializeMethod(Symbol kSym) {
      */
     JCNewClass makeNewClass(Type ctype, List<JCExpression> args, Symbol cons) {
         JCNewClass tree = make.NewClass(null,
-            null, make.QualIdent(ctype.tsym), args, null);
+                null, make.QualIdent(ctype.tsym), args, null);
         tree.constructor = cons;
         tree.type = ctype;
         return tree;
@@ -680,7 +691,7 @@ JCNewClass makeNewClass(Type ctype, List<JCExpression> args, Symbol cons) {
     JCNewClass makeNewClass(Type ctype, List<JCExpression> args) {
         return makeNewClass(ctype, args,
                 rs.resolveConstructor(null, attrEnv, ctype, TreeInfo.types(args), List.nil()));
-     }
+    }
 
     private void addDeserializationCase(MethodHandleSymbol refSym, Type targetType, MethodSymbol samSym,
                                         DiagnosticPosition pos, List<LoadableConstant> staticArgs, MethodType indyType) {
@@ -711,17 +722,17 @@ private void addDeserializationCase(MethodHandleSymbol refSym, Type targetType,
         }
         JCStatement stmt = make.If(
                 deserTest(deserTest(deserTest(deserTest(deserTest(
-                    kindTest,
-                    "getFunctionalInterfaceClass", functionalInterfaceClass),
-                    "getFunctionalInterfaceMethodName", functionalInterfaceMethodName),
-                    "getFunctionalInterfaceMethodSignature", functionalInterfaceMethodSignature),
-                    "getImplClass", implClass),
-                    "getImplMethodSignature", implMethodSignature),
+                                                        kindTest,
+                                                        "getFunctionalInterfaceClass", functionalInterfaceClass),
+                                                "getFunctionalInterfaceMethodName", functionalInterfaceMethodName),
+                                        "getFunctionalInterfaceMethodSignature", functionalInterfaceMethodSignature),
+                                "getImplClass", implClass),
+                        "getImplMethodSignature", implMethodSignature),
                 make.Return(makeIndyCall(
-                    pos,
-                    syms.lambdaMetafactory,
-                    names.altMetafactory,
-                    staticArgs, indyType, serArgs.toList(), samSym.name)),
+                        pos,
+                        syms.lambdaMetafactory,
+                        names.altMetafactory,
+                        staticArgs, indyType, serArgs.toList(), samSym.name)),
                 null);
         ListBuffer<JCStatement> stmts = kInfo.deserializeCases.get(implMethodName);
         if (stmts == null) {
@@ -742,8 +753,8 @@ private void addDeserializationCase(MethodHandleSymbol refSym, Type targetType,
     }
 
     private JCExpression eqTest(Type argType, JCExpression arg1, JCExpression arg2) {
-        JCBinary testExpr = make.Binary(JCTree.Tag.EQ, arg1, arg2);
-        testExpr.operator = operators.resolveBinary(testExpr, JCTree.Tag.EQ, argType, argType);
+        JCBinary testExpr = make.Binary(Tag.EQ, arg1, arg2);
+        testExpr.operator = operators.resolveBinary(testExpr, Tag.EQ, argType, argType);
         testExpr.setType(syms.booleanType);
         return testExpr;
     }
@@ -756,8 +767,8 @@ private JCExpression deserTest(JCExpression prev, String func, String lit) {
                 make.Select(deserGetter(func, syms.stringType), eqsym).setType(eqmt),
                 List.of(make.Literal(lit)));
         eqtest.setType(syms.booleanType);
-        JCBinary compound = make.Binary(JCTree.Tag.AND, prev, eqtest);
-        compound.operator = operators.resolveBinary(compound, JCTree.Tag.AND, syms.booleanType, syms.booleanType);
+        JCBinary compound = make.Binary(Tag.AND, prev, eqtest);
+        compound.operator = operators.resolveBinary(compound, Tag.AND, syms.booleanType, syms.booleanType);
         compound.setType(syms.booleanType);
         return compound;
     }
@@ -770,9 +781,9 @@ private JCExpression deserGetter(String func, Type type, List<Type> argTypes, Li
         MethodType getmt = new MethodType(argTypes, type, List.nil(), syms.methodClass);
         Symbol getsym = rs.resolveQualifiedMethod(null, attrEnv, syms.serializedLambdaType, names.fromString(func), argTypes, List.nil());
         return make.Apply(
-                    List.nil(),
-                    make.Select(make.Ident(kInfo.deserParamSym).setType(syms.serializedLambdaType), getsym).setType(getmt),
-                    args).setType(type);
+                List.nil(),
+                make.Select(make.Ident(kInfo.deserParamSym).setType(syms.serializedLambdaType), getsym).setType(getmt),
+                args).setType(type);
     }
 
     /**
@@ -782,29 +793,20 @@ private MethodSymbol makePrivateSyntheticMethod(long flags, Name name, Type type
         return new MethodSymbol(flags | SYNTHETIC | PRIVATE, name, type, owner);
     }
 
-    /**
-     * Create new synthetic variable with given flags, name, type, owner
-     */
-    private VarSymbol makeSyntheticVar(long flags, Name name, Type type, Symbol owner) {
-        return new VarSymbol(flags | SYNTHETIC, name, type, owner);
-    }
-
-    // </editor-fold>
-
     private MethodType typeToMethodType(Type mt) {
         Type type = types.erasure(mt);
         return new MethodType(type.getParameterTypes(),
-                        type.getReturnType(),
-                        type.getThrownTypes(),
-                        syms.methodClass);
+                type.getReturnType(),
+                type.getThrownTypes(),
+                syms.methodClass);
     }
 
     /**
      * Generate an indy method call to the meta factory
      */
-    private JCExpression makeMetafactoryIndyCall(TranslationContext<?> context,
-            MethodHandleSymbol refSym, List<JCExpression> indy_args) {
-        JCFunctionalExpression tree = context.tree;
+    private JCExpression makeMetafactoryIndyCall(JCFunctionalExpression tree,
+                                                 MethodHandleSymbol refSym, MethodSymbol nonDedupedRefSym,
+                                                 List<JCExpression> indy_args) {
         //determine the static bsm args
         MethodSymbol samSym = (MethodSymbol) types.findDescriptorSymbol(tree.target.tsym);
         List<LoadableConstant> staticArgs = List.of(
@@ -824,10 +826,17 @@ private JCExpression makeMetafactoryIndyCall(TranslationContext<?> context,
                 List.nil(),
                 syms.methodClass);
 
-        Name metafactoryName = context.needsAltMetafactory() ?
+        List<Symbol> bridges = bridges(tree);
+        boolean isSerializable = isSerializable(tree);
+        boolean needsAltMetafactory = tree.target.isIntersection() ||
+                isSerializable || bridges.length() > 1;
+
+        dumpStats(tree, needsAltMetafactory, nonDedupedRefSym);
+
+        Name metafactoryName = needsAltMetafactory ?
                 names.altMetafactory : names.metafactory;
 
-        if (context.needsAltMetafactory()) {
+        if (needsAltMetafactory) {
             ListBuffer<Type> markers = new ListBuffer<>();
             List<Type> targets = tree.target.isIntersection() ?
                     types.directSupertypes(tree.target) :
@@ -835,14 +844,14 @@ private JCExpression makeMetafactoryIndyCall(TranslationContext<?> context,
             for (Type t : targets) {
                 t = types.erasure(t);
                 if (t.tsym != syms.serializableType.tsym &&
-                    t.tsym != tree.type.tsym &&
-                    t.tsym != syms.objectType.tsym) {
+                        t.tsym != tree.type.tsym &&
+                        t.tsym != syms.objectType.tsym) {
                     markers.append(t);
                 }
             }
-            int flags = context.isSerializable() ? FLAG_SERIALIZABLE : 0;
+            int flags = isSerializable ? FLAG_SERIALIZABLE : 0;
             boolean hasMarkers = markers.nonEmpty();
-            boolean hasBridges = context.bridges.nonEmpty();
+            boolean hasBridges = bridges.nonEmpty();
             if (hasMarkers) {
                 flags |= FLAG_MARKERS;
             }
@@ -855,15 +864,15 @@ private JCExpression makeMetafactoryIndyCall(TranslationContext<?> context,
                 staticArgs = staticArgs.appendList(List.convert(LoadableConstant.class, markers.toList()));
             }
             if (hasBridges) {
-                staticArgs = staticArgs.append(LoadableConstant.Int(context.bridges.length() - 1));
-                for (Symbol s : context.bridges) {
+                staticArgs = staticArgs.append(LoadableConstant.Int(bridges.length() - 1));
+                for (Symbol s : bridges) {
                     Type s_erasure = s.erasure(types);
                     if (!types.isSameType(s_erasure, samSym.erasure(types))) {
                         staticArgs = staticArgs.append(((MethodType)s.erasure(types)));
                     }
                 }
             }
-            if (context.isSerializable()) {
+            if (isSerializable) {
                 int prevPos = make.pos;
                 try {
                     make.at(kInfo.clazz);
@@ -889,18 +898,18 @@ private JCExpression makeIndyCall(DiagnosticPosition pos, Type site, Name bsmNam
         try {
             make.at(pos);
             List<Type> bsm_staticArgs = List.of(syms.methodHandleLookupType,
-                syms.stringType,
-                syms.methodTypeType).appendList(staticArgs.map(types::constantType));
+                    syms.stringType,
+                    syms.methodTypeType).appendList(staticArgs.map(types::constantType));
 
             MethodSymbol bsm = rs.resolveInternalMethod(pos, attrEnv, site,
                     bsmName, bsm_staticArgs, List.nil());
 
             DynamicMethodSymbol dynSym =
                     new DynamicMethodSymbol(methName,
-                                            syms.noSymbol,
-                                            bsm.asHandle(),
-                                            indyType,
-                                            staticArgs.toArray(new LoadableConstant[staticArgs.length()]));
+                            syms.noSymbol,
+                            bsm.asHandle(),
+                            indyType,
+                            staticArgs.toArray(new LoadableConstant[staticArgs.length()]));
             JCFieldAccess qualifier = make.Select(make.QualIdent(site.tsym), bsmName);
             DynamicMethodSymbol existing = kInfo.dynMethSyms.putIfAbsent(
                     dynSym.poolKey(types), dynSym);
@@ -915,833 +924,337 @@ private JCExpression makeIndyCall(DiagnosticPosition pos, Type site, Name bsmNam
         }
     }
 
-    // <editor-fold defaultstate="collapsed" desc="Lambda/reference analyzer">
-    /**
-     * This visitor collects information about translation of a lambda expression.
-     * More specifically, it keeps track of the enclosing contexts and captured locals
-     * accessed by the lambda being translated (as well as other useful info).
-     * It also translates away problems for LambdaToMethod.
-     */
-    class LambdaAnalyzerPreprocessor extends TreeTranslator {
-
-        /** the frame stack - used to reconstruct translation info about enclosing scopes */
-        private List<Frame> frameStack;
+    List<Symbol> bridges(JCFunctionalExpression tree) {
+        ClassSymbol csym =
+                types.makeFunctionalInterfaceClass(attrEnv, names.empty, tree.target, ABSTRACT | INTERFACE);
+        return types.functionalInterfaceBridges(csym);
+    }
 
-        /**
-         * keep the count of lambda expression (used to generate unambiguous
-         * names)
-         */
-        private int lambdaCount = 0;
+    /** does this functional expression require serialization support? */
+    boolean isSerializable(JCFunctionalExpression tree) {
+        if (forceSerializable) {
+            return true;
+        }
+        return types.asSuper(tree.target, syms.serializableType.tsym) != null;
+    }
 
-        /**
-         * keep the count of lambda expression defined in given context (used to
-         * generate unambiguous names for serializable lambdas)
-         */
-        private class SyntheticMethodNameCounter {
-            private Map<String, Integer> map = new HashMap<>();
-            int getIndex(StringBuilder buf) {
-                String temp = buf.toString();
-                Integer count = map.get(temp);
-                if (count == null) {
-                    count = 0;
-                }
-                ++count;
-                map.put(temp, count);
-                return count;
+    void dumpStats(JCFunctionalExpression tree, boolean needsAltMetafactory, Symbol sym) {
+        if (dumpLambdaToMethodStats) {
+            if (tree instanceof JCLambda lambda) {
+                log.note(tree, diags.noteKey(lambda.wasMethodReference ? "mref.stat.1" : "lambda.stat",
+                        needsAltMetafactory, sym));
+            } else if (tree instanceof JCMemberReference) {
+                log.note(tree, Notes.MrefStat(needsAltMetafactory, null));
             }
         }
-        private SyntheticMethodNameCounter syntheticMethodNameCounts =
-                new SyntheticMethodNameCounter();
+    }
 
-        private Map<Symbol, JCClassDecl> localClassDefs;
+    /**
+     * This class retains all the useful information about a lambda expression,
+     * and acts as a translation map that is used by the main translation routines
+     * in order to adjust references to captured locals/members, etc.
+     */
+    class LambdaTranslationContext {
 
-        /**
-         * maps for fake clinit symbols to be used as owners of lambda occurring in
-         * a static var init context
-         */
-        private Map<ClassSymbol, Symbol> clinits = new HashMap<>();
+        /** the underlying (untranslated) tree */
+        final JCFunctionalExpression tree;
 
-        private JCClassDecl analyzeAndPreprocessClass(JCClassDecl tree) {
-            frameStack = List.nil();
-            localClassDefs = new HashMap<>();
-            return translate(tree);
-        }
+        /** a translation map from source symbols to translated symbols */
+        final Map<VarSymbol, VarSymbol> lambdaProxies = new HashMap<>();
 
-        @Override
-        public void visitBlock(JCBlock tree) {
-            List<Frame> prevStack = frameStack;
-            try {
-                if (frameStack.nonEmpty() && frameStack.head.tree.hasTag(CLASSDEF)) {
-                    frameStack = frameStack.prepend(new Frame(tree));
-                }
-                super.visitBlock(tree);
-            }
-            finally {
-                frameStack = prevStack;
-            }
-        }
+        /** the list of symbols captured by this lambda expression */
+        final List<VarSymbol> capturedVars;
 
-        @Override
-        public void visitClassDef(JCClassDecl tree) {
-            List<Frame> prevStack = frameStack;
-            int prevLambdaCount = lambdaCount;
-            SyntheticMethodNameCounter prevSyntheticMethodNameCounts =
-                    syntheticMethodNameCounts;
-            Map<ClassSymbol, Symbol> prevClinits = clinits;
-            DiagnosticSource prevSource = log.currentSource();
-            try {
-                log.useSource(tree.sym.sourcefile);
-                lambdaCount = 0;
-                syntheticMethodNameCounts = new SyntheticMethodNameCounter();
-                prevClinits = new HashMap<>();
-                if (tree.sym.owner.kind == MTH) {
-                    localClassDefs.put(tree.sym, tree);
-                }
-                if (directlyEnclosingLambda() != null) {
-                    tree.sym.owner = owner();
-                }
-                frameStack = frameStack.prepend(new Frame(tree));
-                super.visitClassDef(tree);
-            }
-            finally {
-                log.useSource(prevSource.getFile());
-                frameStack = prevStack;
-                lambdaCount = prevLambdaCount;
-                syntheticMethodNameCounts = prevSyntheticMethodNameCounts;
-                clinits = prevClinits;
-            }
-        }
+        /** the synthetic symbol for the method hoisting the translated lambda */
+        final MethodSymbol translatedSym;
 
-        @Override
-        public void visitIdent(JCIdent tree) {
-            if (context() != null && lambdaIdentSymbolFilter(tree.sym)) {
-                if (tree.sym.kind == VAR &&
-                        tree.sym.owner.kind == MTH &&
-                        tree.type.constValue() == null) {
-                    TranslationContext<?> localContext = context();
-                    while (localContext != null) {
-                        if (localContext.tree.getTag() == LAMBDA) {
-                            JCTree block = capturedDecl(localContext.depth, tree.sym);
-                            if (block == null) break;
-                            ((LambdaTranslationContext)localContext)
-                                    .addSymbol(tree.sym, CAPTURED_VAR);
-                        }
-                        localContext = localContext.prev;
-                    }
-                } else if (tree.sym.owner.kind == TYP) {
-                    TranslationContext<?> localContext = context();
-                    while (localContext != null  && !localContext.owner.isStatic()) {
-                        if (localContext.tree.hasTag(LAMBDA)) {
-                            JCTree block = capturedDecl(localContext.depth, tree.sym);
-                            if (block == null) break;
-                            switch (block.getTag()) {
-                                case CLASSDEF:
-                                    JCClassDecl cdecl = (JCClassDecl)block;
-                                    ((LambdaTranslationContext)localContext)
-                                            .addSymbol(cdecl.sym, CAPTURED_THIS);
-                                    break;
-                                default:
-                                    Assert.error("bad block kind");
-                            }
-                        }
-                        localContext = localContext.prev;
+        /** the list of parameter declarations of the translated lambda method */
+        final List<JCVariableDecl> syntheticParams;
+
+        LambdaTranslationContext(JCLambda tree) {
+            this.tree = tree;
+            // This symbol will be filled-in in complete
+            Symbol owner = tree.owner;
+            if (owner.kind == MTH) {
+                final MethodSymbol originalOwner = (MethodSymbol)owner.clone(owner.owner);
+                this.translatedSym = new MethodSymbol(0, null, null, owner.enclClass()) {
+                    @Override
+                    public MethodSymbol originalEnclosingMethod() {
+                        return originalOwner;
                     }
-                }
+                };
+            } else {
+                this.translatedSym = makePrivateSyntheticMethod(0, null, null, owner.enclClass());
             }
-            super.visitIdent(tree);
-        }
-
-        @Override
-        public void visitLambda(JCLambda tree) {
-            analyzeLambda(tree, tree.wasMethodReference ? "mref.stat.1" : "lambda.stat");
-        }
-
-        private LambdaTranslationContext analyzeLambda(JCLambda tree, String statKey) {
-            List<Frame> prevStack = frameStack;
-            try {
-                LambdaTranslationContext context = new LambdaTranslationContext(tree);
-                frameStack = frameStack.prepend(new Frame(tree));
-                for (JCVariableDecl param : tree.params) {
-                    context.addSymbol(param.sym, PARAM);
-                    frameStack.head.addLocal(param.sym);
-                }
-                contextMap.put(tree, context);
-                super.visitLambda(tree);
-                context.complete();
-                if (dumpLambdaToMethodStats) {
-                    log.note(tree, diags.noteKey(statKey, context.needsAltMetafactory(), context.translatedSym));
-                }
-                return context;
+            ListBuffer<JCVariableDecl> params = new ListBuffer<>();
+            ListBuffer<VarSymbol> parameterSymbols = new ListBuffer<>();
+            LambdaCaptureScanner captureScanner = new LambdaCaptureScanner(tree);
+            capturedVars = captureScanner.analyzeCaptures();
+            for (VarSymbol captured : capturedVars) {
+                VarSymbol trans = addSymbol(captured, LambdaSymbolKind.CAPTURED_VAR);
+                params.append(make.VarDef(trans, null));
+                parameterSymbols.add(trans);
             }
-            finally {
-                frameStack = prevStack;
+            for (JCVariableDecl param : tree.params) {
+                VarSymbol trans = addSymbol(param.sym, LambdaSymbolKind.PARAM);
+                params.append(make.VarDef(trans, null));
+                parameterSymbols.add(trans);
             }
+            syntheticParams = params.toList();
+            completeLambdaMethodSymbol(owner, captureScanner.capturesThis);
+            translatedSym.params = parameterSymbols.toList();
         }
 
-        @Override
-        public void visitMethodDef(JCMethodDecl tree) {
-            List<Frame> prevStack = frameStack;
-            try {
-                frameStack = frameStack.prepend(new Frame(tree));
-                super.visitMethodDef(tree);
-            }
-            finally {
-                frameStack = prevStack;
-            }
+        void completeLambdaMethodSymbol(Symbol owner, boolean thisReferenced) {
+            boolean inInterface = owner.enclClass().isInterface();
+
+            // Compute and set the lambda name
+            Name name = isSerializable(tree)
+                    ? serializedLambdaName(owner)
+                    : lambdaName(owner);
+
+            //prepend synthetic args to translated lambda method signature
+            Type type = types.createMethodTypeWithParameters(
+                    generatedLambdaSig(),
+                    TreeInfo.types(syntheticParams));
+
+            // If instance access isn't needed, make it static.
+            // Interface instance methods must be default methods.
+            // Lambda methods are private synthetic.
+            // Inherit ACC_STRICT from the enclosing method, or, for clinit,
+            // from the class.
+            long flags = SYNTHETIC | LAMBDA_METHOD |
+                    owner.flags_field & STRICTFP |
+                    owner.owner.flags_field & STRICTFP |
+                    PRIVATE |
+                    (thisReferenced? (inInterface? DEFAULT : 0) : STATIC);
+
+            translatedSym.type = type;
+            translatedSym.name = name;
+            translatedSym.flags_field = flags;
         }
 
         /**
-         * Method references to local class constructors, may, if the local
-         * class references local variables, have implicit constructor
-         * parameters added in Lower; As a result, the invokedynamic bootstrap
-         * information added in the LambdaToMethod pass will have the wrong
-         * signature. Hooks between Lower and LambdaToMethod have been added to
-         * handle normal "new" in this case. This visitor converts potentially
-         * affected method references into a lambda containing a normal
-         * expression.
+         * For a serializable lambda, generate a disambiguating string
+         * which maximizes stability across deserialization.
          *
-         * @param tree
+         * @return String to differentiate synthetic lambda method names
          */
-        @Override
-        public void visitReference(JCMemberReference tree) {
-            ReferenceTranslationContext rcontext = new ReferenceTranslationContext(tree);
-            contextMap.put(tree, rcontext);
-            super.visitReference(tree);
-            if (dumpLambdaToMethodStats) {
-                log.note(tree, Notes.MrefStat(rcontext.needsAltMetafactory(), null));
+        private String serializedLambdaDisambiguation(Symbol owner) {
+            StringBuilder buf = new StringBuilder();
+            // Append the enclosing method signature to differentiate
+            // overloaded enclosing methods.  For lambdas enclosed in
+            // lambdas, the generated lambda method will not have type yet,
+            // but the enclosing method's name will have been generated
+            // with this same method, so it will be unique and never be
+            // overloaded.
+            Assert.check(
+                    owner.type != null ||
+                            lambdaContext != null);
+            if (owner.type != null) {
+                buf.append(typeSig(owner.type, true));
+                buf.append(":");
             }
-        }
 
-        @Override
-        public void visitSelect(JCFieldAccess tree) {
-            if (context() != null && tree.sym.kind == VAR &&
-                        (tree.sym.name == names._this ||
-                         tree.sym.name == names._super)) {
-                // A select of this or super means, if we are in a lambda,
-                // we much have an instance context
-                TranslationContext<?> localContext = context();
-                while (localContext != null  && !localContext.owner.isStatic()) {
-                    if (localContext.tree.hasTag(LAMBDA)) {
-                        JCClassDecl clazz = (JCClassDecl)capturedDecl(localContext.depth, tree.sym);
-                        if (clazz == null) break;
-                        ((LambdaTranslationContext)localContext).addSymbol(clazz.sym, CAPTURED_THIS);
-                    }
-                    localContext = localContext.prev;
-                }
-            }
-            super.visitSelect(tree);
-        }
+            // Add target type info
+            buf.append(types.findDescriptorSymbol(tree.type.tsym).owner.flatName());
+            buf.append(" ");
 
-        @Override
-        public void visitVarDef(JCVariableDecl tree) {
-            TranslationContext<?> context = context();
-            if (context != null && context instanceof LambdaTranslationContext lambdaContext) {
-                for (Frame frame : frameStack) {
-                    if (frame.tree.hasTag(VARDEF)) {
-                        //skip variable frames inside a lambda:
-                        continue;
-                    } else if (frame.tree.hasTag(LAMBDA)) {
-                        lambdaContext.addSymbol(tree.sym, LOCAL_VAR);
-                    } else {
-                        break;
-                    }
-                }
-                // Check for type variables (including as type arguments).
-                // If they occur within class nested in a lambda, mark for erasure
-                Type type = tree.sym.asType();
+            // Add variable assigned to
+            if (pendingVar != null) {
+                buf.append(pendingVar.flatName());
+                buf.append("=");
             }
-
-            List<Frame> prevStack = frameStack;
-            try {
-                if (tree.sym.owner.kind == MTH) {
-                    frameStack.head.addLocal(tree.sym);
+            //add captured locals info: type, name, order
+            for (Symbol fv : capturedVars) {
+                if (fv != owner) {
+                    buf.append(typeSig(fv.type, true));
+                    buf.append(" ");
+                    buf.append(fv.flatName());
+                    buf.append(",");
                 }
-                frameStack = frameStack.prepend(new Frame(tree));
-                super.visitVarDef(tree);
-            }
-            finally {
-                frameStack = prevStack;
             }
+
+            return buf.toString();
         }
 
         /**
-         * Return a valid owner given the current declaration stack
-         * (required to skip synthetic lambda symbols)
+         * For a non-serializable lambda, generate a simple method.
+         *
+         * @return Name to use for the synthetic lambda method name
          */
-        private Symbol owner() {
-            return owner(false);
+        private Name lambdaName(Symbol owner) {
+            StringBuilder buf = new StringBuilder();
+            buf.append(names.lambda);
+            buf.append(syntheticMethodNameComponent(owner));
+            buf.append("$");
+            buf.append(kInfo.syntheticNameIndex(buf, 0));
+            return names.fromString(buf.toString());
         }
 
-        @SuppressWarnings("fallthrough")
-        private Symbol owner(boolean skipLambda) {
-            List<Frame> frameStack2 = frameStack;
-            while (frameStack2.nonEmpty()) {
-                switch (frameStack2.head.tree.getTag()) {
-                    case VARDEF:
-                        if (((JCVariableDecl)frameStack2.head.tree).sym.isDirectlyOrIndirectlyLocal()) {
-                            frameStack2 = frameStack2.tail;
-                            break;
-                        }
-                        JCClassDecl cdecl = (JCClassDecl)frameStack2.tail.head.tree;
-                        return initSym(cdecl.sym,
-                                ((JCVariableDecl)frameStack2.head.tree).sym.flags() & STATIC);
-                    case BLOCK:
-                        JCClassDecl cdecl2 = (JCClassDecl)frameStack2.tail.head.tree;
-                        return initSym(cdecl2.sym,
-                                ((JCBlock)frameStack2.head.tree).flags & STATIC);
-                    case CLASSDEF:
-                        return ((JCClassDecl)frameStack2.head.tree).sym;
-                    case METHODDEF:
-                        return ((JCMethodDecl)frameStack2.head.tree).sym;
-                    case LAMBDA:
-                        if (!skipLambda)
-                            return ((LambdaTranslationContext)contextMap
-                                    .get(frameStack2.head.tree)).translatedSym;
-                    default:
-                        frameStack2 = frameStack2.tail;
-                }
-            }
-            Assert.error();
-            return null;
-        }
-
-        private Symbol initSym(ClassSymbol csym, long flags) {
-            boolean isStatic = (flags & STATIC) != 0;
-            if (isStatic) {
-                /* static clinits are generated in Gen, so we need to use a fake
-                 * one. Attr creates a fake clinit method while attributing
-                 * lambda expressions used as initializers of static fields, so
-                 * let's use that one.
-                 */
-                MethodSymbol clinit = attr.removeClinit(csym);
-                if (clinit != null) {
-                    clinits.put(csym, clinit);
-                    return clinit;
-                }
-
-                /* if no clinit is found at Attr, then let's try at clinits.
-                 */
-                clinit = (MethodSymbol)clinits.get(csym);
-                if (clinit == null) {
-                    /* no luck, let's create a new one
-                     */
-                    clinit = makePrivateSyntheticMethod(STATIC,
-                            names.clinit,
-                            new MethodType(List.nil(), syms.voidType,
-                                List.nil(), syms.methodClass),
-                            csym);
-                    clinits.put(csym, clinit);
-                }
-                return clinit;
+        /**
+         * @return Method name in a form that can be folded into a
+         * component of a synthetic method name
+         */
+        String syntheticMethodNameComponent(Symbol owner) {
+            long ownerFlags = owner.flags();
+            if ((ownerFlags & BLOCK) != 0) {
+                return (ownerFlags & STATIC) != 0 ?
+                        "static" : "new";
+            } else if (owner.isConstructor()) {
+                return "new";
             } else {
-                //get the first constructor and treat it as the instance init sym
-                for (Symbol s : csym.members_field.getSymbolsByName(names.init)) {
-                    return s;
-                }
+                return owner.name.toString();
             }
-            Assert.error("init not found");
-            return null;
-        }
-
-        private JCTree directlyEnclosingLambda() {
-            if (frameStack.isEmpty()) {
-                return null;
-            }
-            List<Frame> frameStack2 = frameStack;
-            while (frameStack2.nonEmpty()) {
-                switch (frameStack2.head.tree.getTag()) {
-                    case CLASSDEF:
-                    case METHODDEF:
-                        return null;
-                    case LAMBDA:
-                        return frameStack2.head.tree;
-                    default:
-                        frameStack2 = frameStack2.tail;
-                }
-            }
-            Assert.error();
-            return null;
-        }
-
-        private boolean inClassWithinLambda() {
-            if (frameStack.isEmpty()) {
-                return false;
-            }
-            List<Frame> frameStack2 = frameStack;
-            boolean classFound = false;
-            while (frameStack2.nonEmpty()) {
-                switch (frameStack2.head.tree.getTag()) {
-                    case LAMBDA:
-                        return classFound;
-                    case CLASSDEF:
-                        classFound = true;
-                        frameStack2 = frameStack2.tail;
-                        break;
-                    default:
-                        frameStack2 = frameStack2.tail;
-                }
-            }
-            // No lambda
-            return false;
         }
 
         /**
-         * Return the declaration corresponding to a symbol in the enclosing
-         * scope; the depth parameter is used to filter out symbols defined
-         * in nested scopes (which do not need to undergo capture).
+         * For a serializable lambda, generate a method name which maximizes
+         * name stability across deserialization.
+         *
+         * @return Name to use for the synthetic lambda method name
          */
-        private JCTree capturedDecl(int depth, Symbol sym) {
-            Assert.check(sym.kind != TYP);
-            int currentDepth = frameStack.size() - 1;
-            for (Frame block : frameStack) {
-                switch (block.tree.getTag()) {
-                    case CLASSDEF:
-                        ClassSymbol clazz = ((JCClassDecl)block.tree).sym;
-                        if (clazz.isSubClass(sym.enclClass(), types)) {
-                            return currentDepth > depth ? null : block.tree;
-                        }
-                        break;
-                    case VARDEF:
-                        if ((((JCVariableDecl)block.tree).sym == sym &&
-                                sym.owner.kind == MTH) || //only locals are captured
-                            (block.locals != null && block.locals.contains(sym))) {
-                            return currentDepth > depth ? null : block.tree;
-                        }
-                        break;
-                    case BLOCK:
-                    case METHODDEF:
-                    case LAMBDA:
-                        if (block.locals != null && block.locals.contains(sym)) {
-                            return currentDepth > depth ? null : block.tree;
-                        }
-                        break;
-                    default:
-                        Assert.error("bad decl kind " + block.tree.getTag());
-                }
-                currentDepth--;
-            }
-            return null;
-        }
-
-        private TranslationContext<?> context() {
-            for (Frame frame : frameStack) {
-                TranslationContext<?> context = contextMap.get(frame.tree);
-                if (context != null) {
-                    return context;
-                }
-            }
-            return null;
+        private Name serializedLambdaName(Symbol owner) {
+            StringBuilder buf = new StringBuilder();
+            buf.append(names.lambda);
+            // Append the name of the method enclosing the lambda.
+            buf.append(syntheticMethodNameComponent(owner));
+            buf.append('$');
+            // Append a hash of the disambiguating string : enclosing method
+            // signature, etc.
+            String disam = serializedLambdaDisambiguation(owner);
+            buf.append(Integer.toHexString(disam.hashCode()));
+            buf.append('$');
+            // The above appended name components may not be unique, append
+            // a count based on the above name components.
+            buf.append(kInfo.syntheticNameIndex(buf, 1));
+            String result = buf.toString();
+            //System.err.printf("serializedLambdaName: %s -- %s\n", result, disam);
+            return names.fromString(result);
         }
 
         /**
-         *  This is used to filter out those identifiers that needs to be adjusted
-         *  when translating away lambda expressions
+         * Translate a symbol of a given kind into something suitable for the
+         * synthetic lambda body
          */
-        private boolean lambdaIdentSymbolFilter(Symbol sym) {
-            return (sym.kind == VAR || sym.kind == MTH)
-                    && !sym.isStatic()
-                    && sym.name != names.init;
-        }
-
-        private class Frame {
-            final JCTree tree;
-            List<Symbol> locals;
-
-            public Frame(JCTree tree) {
-                this.tree = tree;
+        VarSymbol translate(final VarSymbol sym, LambdaSymbolKind skind) {
+            VarSymbol ret;
+            boolean propagateAnnos = true;
+            switch (skind) {
+                case CAPTURED_VAR:
+                    Name name = (sym.flags() & LOCAL_CAPTURE_FIELD) != 0 ?
+                            sym.baseSymbol().name : sym.name;
+                    ret = new VarSymbol(SYNTHETIC | FINAL | PARAMETER, name, types.erasure(sym.type), translatedSym);
+                    propagateAnnos = false;
+                    break;
+                case LOCAL_VAR:
+                    ret = new VarSymbol(sym.flags() & FINAL, sym.name, sym.type, translatedSym);
+                    ret.pos = sym.pos;
+                    // If sym.data == ElementKind.EXCEPTION_PARAMETER,
+                    // set ret.data = ElementKind.EXCEPTION_PARAMETER too.
+                    // Because method com.sun.tools.javac.jvm.Code.fillExceptionParameterPositions and
+                    // com.sun.tools.javac.jvm.Code.fillLocalVarPosition would use it.
+                    // See JDK-8257740 for more information.
+                    if (sym.isExceptionParameter()) {
+                        ret.setData(ElementKind.EXCEPTION_PARAMETER);
+                    }
+                    break;
+                case PARAM:
+                    ret = new VarSymbol((sym.flags() & FINAL) | PARAMETER, sym.name, types.erasure(sym.type), translatedSym);
+                    ret.pos = sym.pos;
+                    break;
+                default:
+                    Assert.error(skind.name());
+                    throw new AssertionError();
             }
-
-            void addLocal(Symbol sym) {
-                if (locals == null) {
-                    locals = List.nil();
-                }
-                locals = locals.prepend(sym);
+            if (ret != sym && propagateAnnos) {
+                ret.setDeclarationAttributes(sym.getRawAttributes());
+                ret.setTypeAttributes(sym.getRawTypeAttributes());
             }
+            return ret;
         }
 
-        /**
-         * This class is used to store important information regarding translation of
-         * lambda expression/method references (see subclasses).
-         */
-        abstract class TranslationContext<T extends JCFunctionalExpression> {
-
-            /** the underlying (untranslated) tree */
-            final T tree;
-
-            /** points to the adjusted enclosing scope in which this lambda/mref expression occurs */
-            final Symbol owner;
-
-            /** the depth of this lambda expression in the frame stack */
-            final int depth;
-
-            /** the enclosing translation context (set for nested lambdas/mref) */
-            final TranslationContext<?> prev;
-
-            /** list of methods to be bridged by the meta-factory */
-            final List<Symbol> bridges;
-
-            TranslationContext(T tree) {
-                this.tree = tree;
-                this.owner = owner(true);
-                this.depth = frameStack.size() - 1;
-                this.prev = context();
-                ClassSymbol csym =
-                        types.makeFunctionalInterfaceClass(attrEnv, names.empty, tree.target, ABSTRACT | INTERFACE);
-                this.bridges = types.functionalInterfaceBridges(csym);
-            }
-
-            /** does this functional expression need to be created using alternate metafactory? */
-            boolean needsAltMetafactory() {
-                return tree.target.isIntersection() ||
-                        isSerializable() ||
-                        bridges.length() > 1;
-            }
+        VarSymbol addLocal(VarSymbol sym) {
+            return addSymbol(sym, LambdaSymbolKind.LOCAL_VAR);
+        }
 
-            /** does this functional expression require serialization support? */
-            boolean isSerializable() {
-                if (forceSerializable) {
-                    return true;
-                }
-                return types.asSuper(tree.target, syms.serializableType.tsym) != null;
-            }
+        private VarSymbol addSymbol(VarSymbol sym, LambdaSymbolKind skind) {
+            return lambdaProxies.computeIfAbsent(sym, s -> translate(s, skind));
+        }
 
-            /**
-             * @return Name of the enclosing method to be folded into synthetic
-             * method name
-             */
-            String enclosingMethodName() {
-                return syntheticMethodNameComponent(owner.name);
-            }
+        JCTree translate(JCIdent lambdaIdent) {
+            Symbol tSym = lambdaProxies.get(lambdaIdent.sym);
+            return tSym != null ?
+                    make.Ident(tSym).setType(lambdaIdent.type) :
+                    null;
+        }
 
-            /**
-             * @return Method name in a form that can be folded into a
-             * component of a synthetic method name
-             */
-            String syntheticMethodNameComponent(Name name) {
-                if (name == null) {
-                    return "null";
-                }
-                String methodName = name.toString();
-                if (methodName.equals("<clinit>")) {
-                    methodName = "static";
-                } else if (methodName.equals("<init>")) {
-                    methodName = "new";
-                }
-                return methodName;
-            }
+        Type generatedLambdaSig() {
+            return types.erasure(tree.getDescriptorType(types));
         }
 
         /**
-         * This class retains all the useful information about a lambda expression;
-         * the contents of this class are filled by the LambdaAnalyzer visitor,
-         * and the used by the main translation routines in order to adjust references
-         * to captured locals/members, etc.
+         * Compute the set of local variables captured by this lambda expression.
+         * Also determines whether this lambda expression captures the enclosing 'this'.
          */
-        class LambdaTranslationContext extends TranslationContext<JCLambda> {
-
-            /** variable in the enclosing context to which this lambda is assigned */
-            final Symbol self;
-
-            /** variable in the enclosing context to which this lambda is assigned */
-            final Symbol assignedTo;
-
-            Map<LambdaSymbolKind, Map<Symbol, Symbol>> translatedSymbols;
-
-            /** the synthetic symbol for the method hoisting the translated lambda */
-            MethodSymbol translatedSym;
-
-            List<JCVariableDecl> syntheticParams;
-
-            LambdaTranslationContext(JCLambda tree) {
-                super(tree);
-                Frame frame = frameStack.head;
-                switch (frame.tree.getTag()) {
-                    case VARDEF:
-                        assignedTo = self = ((JCVariableDecl) frame.tree).sym;
-                        break;
-                    case ASSIGN:
-                        self = null;
-                        assignedTo = TreeInfo.symbol(((JCAssign) frame.tree).getVariable());
-                        break;
-                    default:
-                        assignedTo = self = null;
-                        break;
-                 }
-
-                // This symbol will be filled-in in complete
-                if (owner.kind == MTH) {
-                    final MethodSymbol originalOwner = (MethodSymbol)owner.clone(owner.owner);
-                    this.translatedSym = new MethodSymbol(SYNTHETIC | PRIVATE, null, null, owner.enclClass()) {
-                        @Override
-                        public MethodSymbol originalEnclosingMethod() {
-                            return originalOwner;
-                        }
-                    };
-                } else {
-                    this.translatedSym = makePrivateSyntheticMethod(0, null, null, owner.enclClass());
-                }
-                translatedSymbols = new EnumMap<>(LambdaSymbolKind.class);
-
-                translatedSymbols.put(PARAM, new LinkedHashMap<>());
-                translatedSymbols.put(LOCAL_VAR, new LinkedHashMap<>());
-                translatedSymbols.put(CAPTURED_VAR, new LinkedHashMap<>());
-                translatedSymbols.put(CAPTURED_THIS, new LinkedHashMap<>());
-            }
-
-             /**
-             * For a serializable lambda, generate a disambiguating string
-             * which maximizes stability across deserialization.
-             *
-             * @return String to differentiate synthetic lambda method names
-             */
-            private String serializedLambdaDisambiguation() {
-                StringBuilder buf = new StringBuilder();
-                // Append the enclosing method signature to differentiate
-                // overloaded enclosing methods.  For lambdas enclosed in
-                // lambdas, the generated lambda method will not have type yet,
-                // but the enclosing method's name will have been generated
-                // with this same method, so it will be unique and never be
-                // overloaded.
-                Assert.check(
-                        owner.type != null ||
-                        directlyEnclosingLambda() != null);
-                if (owner.type != null) {
-                    buf.append(typeSig(owner.type, true));
-                    buf.append(":");
-                }
-
-                // Add target type info
-                buf.append(types.findDescriptorSymbol(tree.type.tsym).owner.flatName());
-                buf.append(" ");
-
-                // Add variable assigned to
-                if (assignedTo != null) {
-                    buf.append(assignedTo.flatName());
-                    buf.append("=");
-                }
-                //add captured locals info: type, name, order
-                for (Symbol fv : getSymbolMap(CAPTURED_VAR).keySet()) {
-                    if (fv != self) {
-                        buf.append(typeSig(fv.type, true));
-                        buf.append(" ");
-                        buf.append(fv.flatName());
-                        buf.append(",");
-                    }
-                }
-
-                return buf.toString();
-            }
+        class LambdaCaptureScanner extends CaptureScanner {
+            boolean capturesThis;
+            Set<ClassSymbol> seenClasses = new HashSet<>();
 
-            /**
-             * For a non-serializable lambda, generate a simple method.
-             *
-             * @return Name to use for the synthetic lambda method name
-             */
-            private Name lambdaName() {
-                return names.lambda.append(names.fromString(enclosingMethodName() + "$" + lambdaCount++));
+            LambdaCaptureScanner(JCLambda ownerTree) {
+                super(ownerTree);
             }
 
-            /**
-             * For a serializable lambda, generate a method name which maximizes
-             * name stability across deserialization.
-             *
-             * @return Name to use for the synthetic lambda method name
-             */
-            private Name serializedLambdaName() {
-                StringBuilder buf = new StringBuilder();
-                buf.append(names.lambda);
-                // Append the name of the method enclosing the lambda.
-                buf.append(enclosingMethodName());
-                buf.append('$');
-                // Append a hash of the disambiguating string : enclosing method
-                // signature, etc.
-                String disam = serializedLambdaDisambiguation();
-                buf.append(Integer.toHexString(disam.hashCode()));
-                buf.append('$');
-                // The above appended name components may not be unique, append
-                // a count based on the above name components.
-                buf.append(syntheticMethodNameCounts.getIndex(buf));
-                String result = buf.toString();
-                //System.err.printf("serializedLambdaName: %s -- %s\n", result, disam);
-                return names.fromString(result);
-            }
-
-            /**
-             * Translate a symbol of a given kind into something suitable for the
-             * synthetic lambda body
-             */
-            Symbol translate(final Symbol sym, LambdaSymbolKind skind) {
-                Symbol ret;
-                switch (skind) {
-                    case CAPTURED_THIS:
-                        ret = sym;  // self represented
-                        break;
-                    case CAPTURED_VAR:
-                        ret = new VarSymbol(SYNTHETIC | FINAL | PARAMETER, sym.name, types.erasure(sym.type), translatedSym) {
-                            @Override
-                            public Symbol baseSymbol() {
-                                //keep mapping with original captured symbol
-                                return sym;
-                            }
-                        };
-                        break;
-                    case LOCAL_VAR:
-                        ret = new VarSymbol(sym.flags() & FINAL, sym.name, sym.type, translatedSym) {
-                            @Override
-                            public Symbol baseSymbol() {
-                                //keep mapping with original symbol
-                                return sym;
-                            }
-                        };
-                        ((VarSymbol) ret).pos = ((VarSymbol) sym).pos;
-                        // If sym.data == ElementKind.EXCEPTION_PARAMETER,
-                        // set ret.data = ElementKind.EXCEPTION_PARAMETER too.
-                        // Because method com.sun.tools.javac.jvm.Code.fillExceptionParameterPositions and
-                        // com.sun.tools.javac.jvm.Code.fillLocalVarPosition would use it.
-                        // See JDK-8257740 for more information.
-                        if (((VarSymbol) sym).isExceptionParameter()) {
-                            ((VarSymbol) ret).setData(ElementKind.EXCEPTION_PARAMETER);
-                        }
-                        break;
-                    case PARAM:
-                        ret = new VarSymbol((sym.flags() & FINAL) | PARAMETER, sym.name, types.erasure(sym.type), translatedSym);
-                        ((VarSymbol) ret).pos = ((VarSymbol) sym).pos;
-                        // Set ret.data. Same as case LOCAL_VAR above.
-                        if (((VarSymbol) sym).isExceptionParameter()) {
-                            ((VarSymbol) ret).setData(ElementKind.EXCEPTION_PARAMETER);
-                        }
-                        break;
-                    default:
-                        Assert.error(skind.name());
-                        throw new AssertionError();
-                }
-                if (ret != sym && skind.propagateAnnotations()) {
-                    ret.setDeclarationAttributes(sym.getRawAttributes());
-                    ret.setTypeAttributes(sym.getRawTypeAttributes());
-                }
-                return ret;
-            }
-
-            void addSymbol(Symbol sym, LambdaSymbolKind skind) {
-                Map<Symbol, Symbol> transMap = getSymbolMap(skind);
-                if (!transMap.containsKey(sym)) {
-                    transMap.put(sym, translate(sym, skind));
-                }
-            }
-
-            Map<Symbol, Symbol> getSymbolMap(LambdaSymbolKind skind) {
-                Map<Symbol, Symbol> m = translatedSymbols.get(skind);
-                Assert.checkNonNull(m);
-                return m;
+            @Override
+            public void visitClassDef(JCClassDecl tree) {
+                seenClasses.add(tree.sym);
+                super.visitClassDef(tree);
             }
 
-            JCTree translate(JCIdent lambdaIdent) {
-                for (LambdaSymbolKind kind : LambdaSymbolKind.values()) {
-                    Map<Symbol, Symbol> m = getSymbolMap(kind);
-                    switch(kind) {
-                        default:
-                            if (m.containsKey(lambdaIdent.sym)) {
-                                Symbol tSym = m.get(lambdaIdent.sym);
-                                JCTree t = make.Ident(tSym).setType(lambdaIdent.type);
-                                return t;
-                            }
-                            break;
+            @Override
+            public void visitIdent(JCIdent tree) {
+                if (!tree.sym.isStatic() &&
+                        tree.sym.owner.kind == TYP &&
+                        (tree.sym.kind == VAR || tree.sym.kind == MTH) &&
+                        !seenClasses.contains(tree.sym.owner)) {
+                    if ((tree.sym.flags() & LOCAL_CAPTURE_FIELD) != 0) {
+                        // a local, captured by Lower - re-capture!
+                        addFreeVar((VarSymbol) tree.sym);
+                    } else {
+                        // a reference to an enclosing field or method, we need to capture 'this'
+                        capturesThis = true;
                     }
+                } else {
+                    // might be a local capture
+                    super.visitIdent(tree);
                 }
-                return null;
             }
 
-            /**
-             * The translatedSym is not complete/accurate until the analysis is
-             * finished.  Once the analysis is finished, the translatedSym is
-             * "completed" -- updated with type information, access modifiers,
-             * and full parameter list.
-             */
-            void complete() {
-                if (syntheticParams != null) {
-                    return;
-                }
-                boolean inInterface = translatedSym.owner.isInterface();
-                boolean thisReferenced = !getSymbolMap(CAPTURED_THIS).isEmpty();
-
-                // If instance access isn't needed, make it static.
-                // Interface instance methods must be default methods.
-                // Lambda methods are private synthetic.
-                // Inherit ACC_STRICT from the enclosing method, or, for clinit,
-                // from the class.
-                translatedSym.flags_field = SYNTHETIC | LAMBDA_METHOD |
-                        owner.flags_field & STRICTFP |
-                        owner.owner.flags_field & STRICTFP |
-                        PRIVATE |
-                        (thisReferenced? (inInterface? DEFAULT : 0) : STATIC);
-
-                //compute synthetic params
-                ListBuffer<JCVariableDecl> params = new ListBuffer<>();
-                ListBuffer<VarSymbol> parameterSymbols = new ListBuffer<>();
-
-                // The signature of the method is augmented with the following
-                // synthetic parameters:
-                //
-                // 1) reference to enclosing contexts captured by the lambda expression
-                // 2) enclosing locals captured by the lambda expression
-                for (Symbol thisSym : getSymbolMap(CAPTURED_VAR).values()) {
-                    params.append(make.VarDef((VarSymbol) thisSym, null));
-                    parameterSymbols.append((VarSymbol) thisSym);
-                }
-                for (Symbol thisSym : getSymbolMap(PARAM).values()) {
-                    params.append(make.VarDef((VarSymbol) thisSym, null));
-                    parameterSymbols.append((VarSymbol) thisSym);
+            @Override
+            public void visitSelect(JCFieldAccess tree) {
+                if (tree.sym.kind == VAR &&
+                        (tree.sym.name == names._this ||
+                                tree.sym.name == names._super) &&
+                        !seenClasses.contains(tree.sym.type.tsym)) {
+                    capturesThis = true;
                 }
-                syntheticParams = params.toList();
-
-                translatedSym.params = parameterSymbols.toList();
-
-                // Compute and set the lambda name
-                translatedSym.name = isSerializable()
-                        ? serializedLambdaName()
-                        : lambdaName();
-
-                //prepend synthetic args to translated lambda method signature
-                translatedSym.type = types.createMethodTypeWithParameters(
-                        generatedLambdaSig(),
-                        TreeInfo.types(syntheticParams));
+                super.visitSelect(tree);
             }
 
-            Type generatedLambdaSig() {
-                return types.erasure(tree.getDescriptorType(types));
+            @Override
+            public void visitAnnotation(JCAnnotation tree) {
+                // do nothing (annotation values look like captured instance fields)
             }
         }
 
-        /**
-         * Simple subclass modelling the translation context of a method reference.
+        /*
+         * These keys provide mappings for various translated lambda symbols
+         * and the prevailing order must be maintained.
          */
-        final class ReferenceTranslationContext extends TranslationContext<JCMemberReference> {
-
-            ReferenceTranslationContext(JCMemberReference tree) {
-                super(tree);
-            }
-        }
-    }
-    // </editor-fold>
-
-    /*
-     * These keys provide mappings for various translated lambda symbols
-     * and the prevailing order must be maintained.
-     */
-    enum LambdaSymbolKind {
-        PARAM,          // original to translated lambda parameters
-        LOCAL_VAR,      // original to translated lambda locals
-        CAPTURED_VAR,   // variables in enclosing scope to translated synthetic parameters
-        CAPTURED_THIS;  // class symbols to translated synthetic parameters (for captured member access)
-
-        boolean propagateAnnotations() {
-            switch (this) {
-                case CAPTURED_VAR:
-                case CAPTURED_THIS:
-                    return false;
-                default:
-                    return true;
-           }
+        enum LambdaSymbolKind {
+            PARAM,          // original to translated lambda parameters
+            LOCAL_VAR,      // original to translated lambda locals
+            CAPTURED_VAR;   // variables in enclosing scope to translated synthetic parameters
         }
     }
 
@@ -1791,7 +1304,7 @@ private class L2MSignatureGenerator extends Types.SignatureGenerator {
 
         /**
          * Are signatures incompatible with JVM spec allowed?
-         * Used by {@link LambdaTranslationContext#serializedLambdaDisambiguation()}.
+         * Used by {@link LambdaTranslationContext#serializedLambdaDisambiguation(Symbol)}}.
          */
         boolean allowIllegalSignatures;
 
diff --git a/src/jdk.compiler/share/classes/com/sun/tools/javac/comp/Lower.java b/src/jdk.compiler/share/classes/com/sun/tools/javac/comp/Lower.java
index c5fd2177d49a7..62117583a96e9 100644
--- a/src/jdk.compiler/share/classes/com/sun/tools/javac/comp/Lower.java
+++ b/src/jdk.compiler/share/classes/com/sun/tools/javac/comp/Lower.java
@@ -267,26 +267,22 @@ List<Name> enumNamesFor(ClassSymbol c) {
     Map<ClassSymbol,List<VarSymbol>> freevarCache;
 
     /** A navigator class for collecting the free variables accessed
-     *  from a local class. There is only one case; all other cases simply
-     *  traverse down the tree. This class doesn't deal with the specific
-     *  of Lower - it's an abstract visitor that is meant to be reused in
-     *  order to share the local variable capture logic.
+     *  from a local class.
      */
-    abstract class BasicFreeVarCollector extends TreeScanner {
+    class FreeVarCollector extends CaptureScanner {
 
-        /** Add all free variables of class c to fvs list
-         *  unless they are already there.
-         */
-        abstract void addFreeVars(ClassSymbol c);
+        FreeVarCollector(JCTree ownerTree) {
+            super(ownerTree);
+        }
 
-        /** If tree refers to a variable in owner of local class, add it to
-         *  free variables list.
-         */
-        public void visitIdent(JCIdent tree) {
-            visitSymbol(tree.sym);
+        void addFreeVars(ClassSymbol c) {
+            List<VarSymbol> fvs = freevarCache.get(c);
+            if (fvs != null) {
+                for (List<VarSymbol> l = fvs; l.nonEmpty(); l = l.tail) {
+                    addFreeVar(l.head);
+                }
+            }
         }
-        // where
-        abstract void visitSymbol(Symbol _sym);
 
         /** If tree refers to a class instance creation expression
          *  add all free variables of the freshly created class.
@@ -306,84 +302,6 @@ public void visitApply(JCMethodInvocation tree) {
             }
             super.visitApply(tree);
         }
-
-        @Override
-        public void visitYield(JCYield tree) {
-            scan(tree.value);
-        }
-
-    }
-
-    /**
-     * Lower-specific subclass of {@code BasicFreeVarCollector}.
-     */
-    class FreeVarCollector extends BasicFreeVarCollector {
-
-        /** The owner of the local class.
-         */
-        Symbol owner;
-
-        /** The local class.
-         */
-        ClassSymbol clazz;
-
-        /** The list of owner's variables accessed from within the local class,
-         *  without any duplicates.
-         */
-        List<VarSymbol> fvs;
-
-        FreeVarCollector(ClassSymbol clazz) {
-            this.clazz = clazz;
-            this.owner = clazz.owner;
-            this.fvs = List.nil();
-        }
-
-        /** Add free variable to fvs list unless it is already there.
-         */
-        private void addFreeVar(VarSymbol v) {
-            for (List<VarSymbol> l = fvs; l.nonEmpty(); l = l.tail)
-                if (l.head == v) return;
-            fvs = fvs.prepend(v);
-        }
-
-        @Override
-        void addFreeVars(ClassSymbol c) {
-            List<VarSymbol> fvs = freevarCache.get(c);
-            if (fvs != null) {
-                for (List<VarSymbol> l = fvs; l.nonEmpty(); l = l.tail) {
-                    addFreeVar(l.head);
-                }
-            }
-        }
-
-        @Override
-        void visitSymbol(Symbol _sym) {
-            Symbol sym = _sym;
-            if (sym.kind == VAR || sym.kind == MTH) {
-                if (sym != null && sym.owner != owner)
-                    sym = proxies.get(sym);
-                if (sym != null && sym.owner == owner) {
-                    VarSymbol v = (VarSymbol)sym;
-                    if (v.getConstValue() == null) {
-                        addFreeVar(v);
-                    }
-                }
-            }
-        }
-    }
-
-    ClassSymbol ownerToCopyFreeVarsFrom(ClassSymbol c) {
-        if (!c.isDirectlyOrIndirectlyLocal()) {
-            return null;
-        }
-        Symbol currentOwner = c.owner;
-        while (currentOwner.owner.kind.matches(KindSelector.TYP) && currentOwner.isDirectlyOrIndirectlyLocal()) {
-            currentOwner = currentOwner.owner;
-        }
-        if (currentOwner.owner.kind.matches(KindSelector.VAL_MTH) && c.isSubClass(currentOwner, types)) {
-            return (ClassSymbol)currentOwner;
-        }
-        return null;
     }
 
     /** Return the variables accessed from within a local class, which
@@ -395,22 +313,10 @@ List<VarSymbol> freevars(ClassSymbol c)  {
         if (fvs != null) {
             return fvs;
         }
-        if (c.owner.kind.matches(KindSelector.VAL_MTH) && !c.isStatic()) {
-            FreeVarCollector collector = new FreeVarCollector(c);
-            collector.scan(classDef(c));
-            fvs = collector.fvs;
-            freevarCache.put(c, fvs);
-            return fvs;
-        } else {
-            ClassSymbol owner = ownerToCopyFreeVarsFrom(c);
-            if (owner != null) {
-                fvs = freevarCache.get(owner);
-                freevarCache.put(c, fvs);
-                return fvs;
-            } else {
-                return List.nil();
-            }
-        }
+        FreeVarCollector collector = new FreeVarCollector(classDef(c));
+        fvs = collector.analyzeCaptures().reverse();
+        freevarCache.put(c, fvs);
+        return fvs;
     }
 
     Map<TypeSymbol,EnumMapping> enumSwitchMap = new LinkedHashMap<>();
@@ -1501,7 +1407,7 @@ Name proxyName(Name name, int index) {
      *  @param owner      The class in which the definitions go.
      */
     List<JCVariableDecl> freevarDefs(int pos, List<VarSymbol> freevars, Symbol owner) {
-        return freevarDefs(pos, freevars, owner, 0);
+        return freevarDefs(pos, freevars, owner, LOCAL_CAPTURE_FIELD);
     }
 
     List<JCVariableDecl> freevarDefs(int pos, List<VarSymbol> freevars, Symbol owner,
@@ -1517,7 +1423,12 @@ List<JCVariableDecl> freevarDefs(int pos, List<VarSymbol> freevars, Symbol owner
                 proxyName = proxyName(v.name, index++);
             } while (!proxyNames.add(proxyName));
             VarSymbol proxy = new VarSymbol(
-                flags, proxyName, v.erasure(types), owner);
+                flags, proxyName, v.erasure(types), owner) {
+                @Override
+                public Symbol baseSymbol() {
+                    return v;
+                }
+            };
             proxies.put(v, proxy);
             JCVariableDecl vd = make.at(pos).VarDef(proxy, null);
             vd.vartype = access(vd.vartype);
diff --git a/src/jdk.compiler/share/classes/com/sun/tools/javac/comp/TransTypes.java b/src/jdk.compiler/share/classes/com/sun/tools/javac/comp/TransTypes.java
index bfe35dbec9c1f..d232f2e6d9f3d 100644
--- a/src/jdk.compiler/share/classes/com/sun/tools/javac/comp/TransTypes.java
+++ b/src/jdk.compiler/share/classes/com/sun/tools/javac/comp/TransTypes.java
@@ -679,6 +679,7 @@ JCExpression lambda() {
 
                 JCLambda slam = make.Lambda(params.toList(), expr);
                 slam.target = tree.target;
+                slam.owner = tree.owner;
                 slam.type = tree.type;
                 slam.pos = tree.pos;
                 slam.wasMethodReference = true;
diff --git a/src/jdk.compiler/share/classes/com/sun/tools/javac/jvm/ClassWriter.java b/src/jdk.compiler/share/classes/com/sun/tools/javac/jvm/ClassWriter.java
index dfa92efae7421..6679bb43fb807 100644
--- a/src/jdk.compiler/share/classes/com/sun/tools/javac/jvm/ClassWriter.java
+++ b/src/jdk.compiler/share/classes/com/sun/tools/javac/jvm/ClassWriter.java
@@ -329,7 +329,7 @@ protected int writeEnclosingMethodAttribute(Name attributeName, ClassSymbol c) {
         int alenIdx = writeAttr(attributeName);
         ClassSymbol enclClass = c.owner.enclClass();
         MethodSymbol enclMethod =
-            (c.owner.type == null // local to init block
+            ((c.owner.flags() & BLOCK) != 0 // local to init block
              || c.owner.kind != MTH) // or member init
             ? null
             : ((MethodSymbol)c.owner).originalEnclosingMethod();
diff --git a/src/jdk.compiler/share/classes/com/sun/tools/javac/tree/JCTree.java b/src/jdk.compiler/share/classes/com/sun/tools/javac/tree/JCTree.java
index 6f3b8b5d8aaca..6041da6723aa4 100644
--- a/src/jdk.compiler/share/classes/com/sun/tools/javac/tree/JCTree.java
+++ b/src/jdk.compiler/share/classes/com/sun/tools/javac/tree/JCTree.java
@@ -807,6 +807,8 @@ public JCFunctionalExpression() {
 
         /** list of target types inferred for this functional expression. */
         public Type target;
+        /** The owner of this functional expression. */
+        public Symbol owner;
 
         public Type getDescriptorType(Types types) {
             return target != null ? types.findDescriptorType(target) : types.createErrorType(null);
diff --git a/src/jdk.compiler/share/classes/com/sun/tools/javac/tree/TreeScanner.java b/src/jdk.compiler/share/classes/com/sun/tools/javac/tree/TreeScanner.java
index 0336f3c4191ac..b9ae35da9df1f 100644
--- a/src/jdk.compiler/share/classes/com/sun/tools/javac/tree/TreeScanner.java
+++ b/src/jdk.compiler/share/classes/com/sun/tools/javac/tree/TreeScanner.java
@@ -273,8 +273,8 @@ public void visitNewArray(JCNewArray tree) {
     }
 
     public void visitLambda(JCLambda tree) {
-        scan(tree.body);
         scan(tree.params);
+        scan(tree.body);
     }
 
     public void visitParens(JCParens tree) {
diff --git a/src/jdk.hotspot.agent/share/man/jhsdb.1 b/src/jdk.hotspot.agent/share/man/jhsdb.1
index 5b65f7eafb465..93cc1fedb158a 100644
--- a/src/jdk.hotspot.agent/share/man/jhsdb.1
+++ b/src/jdk.hotspot.agent/share/man/jhsdb.1
@@ -1,4 +1,4 @@
-.\" Copyright (c) 2019, 2020, Oracle and/or its affiliates. All rights reserved.
+.\" Copyright (c) 2019, 2024, Oracle and/or its affiliates. All rights reserved.
 .\" DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
 .\"
 .\" This code is free software; you can redistribute it and/or modify it
@@ -44,6 +44,10 @@ analyze the content of a core dump from a crashed Java Virtual Machine
 (JVM)
 .SH SYNOPSIS
 .PP
+\f[B]WARNING:\f[R] The \f[V]debugd\f[R] subcommand and
+\f[V]--connect\f[R] options are deprecated.
+They will be removed in a future release.
+.PP
 \f[V]jhsdb\f[R] \f[V]clhsdb\f[R] [\f[V]--pid\f[R] \f[I]pid\f[R] |
 \f[V]--exe\f[R] \f[I]executable\f[R] \f[V]--core\f[R]
 \f[I]coredump\f[R]]
diff --git a/src/jdk.jpackage/share/man/jpackage.1 b/src/jdk.jpackage/share/man/jpackage.1
index 13d9c41c31d6a..09d340ec03399 100644
--- a/src/jdk.jpackage/share/man/jpackage.1
+++ b/src/jdk.jpackage/share/man/jpackage.1
@@ -111,6 +111,16 @@ Path where generated output file is placed
 Defaults to the current working directory.
 .RE
 .TP
+\f[V]--resource-dir\f[R] \f[I]path\f[R]
+Path to override jpackage resources
+.RS
+.PP
+(absolute path or relative to the current directory)
+.PP
+Icons, template files, and other resources of jpackage can be
+over-ridden by adding replacement resources to this directory.
+.RE
+.TP
 \f[V]--temp\f[R] \f[I]directory\f[R]
 Path of a new or empty directory used to create temporary files
 .RS
@@ -207,9 +217,9 @@ of key, value pairs
 .PP
 The keys \[dq]module\[dq], \[dq]main-jar\[dq], \[dq]main-class\[dq],
 \[dq]description\[dq], \[dq]arguments\[dq], \[dq]java-options\[dq],
-\[dq]app-version\[dq], \[dq]icon\[dq], \[dq]launcher-as-service\[dq],
-\[dq]win-console\[dq], \[dq]win-shortcut\[dq], \[dq]win-menu\[dq],
-\[dq]linux-app-category\[dq], and \[dq]linux-shortcut\[dq] can be used.
+\[dq]icon\[dq], \[dq]launcher-as-service\[dq], \[dq]win-console\[dq],
+\[dq]win-shortcut\[dq], \[dq]win-menu\[dq], and \[dq]linux-shortcut\[dq]
+can be used.
 .PP
 These options are added to, or used to overwrite, the original command
 line options to build an additional alternative launcher.
@@ -260,7 +270,7 @@ When this option is specified, the main module will be linked in the
 Java runtime image.
 Either --module or --main-jar option can be specified but not both.
 .RE
-.SS Platform dependent option for creating the application launcher:
+.SS Platform dependent options for creating the application launcher:
 .SS Windows platform options (available only when running on Windows):
 .TP
 \f[V]--win-console\f[R]
@@ -357,16 +367,6 @@ Path to the license file
 (absolute path or relative to the current directory)
 .RE
 .TP
-\f[V]--resource-dir\f[R] \f[I]path\f[R]
-Path to override jpackage resources
-.RS
-.PP
-(absolute path or relative to the current directory)
-.PP
-Icons, template files, and other resources of jpackage can be
-over-ridden by adding replacement resources to this directory.
-.RE
-.TP
 \f[V]--runtime-image\f[R] \f[I]path\f[R]
 Path of the predefined runtime image to install
 .RS
diff --git a/test/hotspot/jtreg/compiler/floatingpoint/TestRoundFloatAll.java b/test/hotspot/jtreg/compiler/floatingpoint/TestRoundFloatAll.java
new file mode 100644
index 0000000000000..2ba60fd8e319f
--- /dev/null
+++ b/test/hotspot/jtreg/compiler/floatingpoint/TestRoundFloatAll.java
@@ -0,0 +1,84 @@
+/*
+ * Copyright (c) 2024, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2024, Rivos Inc. All rights reserved.
+ * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
+ *
+ * This code is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 only, as
+ * published by the Free Software Foundation.
+ *
+ * This code is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+ * version 2 for more details (a copy is included in the LICENSE file that
+ * accompanied this code).
+ *
+ * You should have received a copy of the GNU General Public License version
+ * 2 along with this work; if not, write to the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
+ * or visit www.oracle.com if you need additional information or have any
+ * questions.
+ */
+
+/**
+ * @test
+ * @bug 8321010
+ * @summary Test intrinsic for Math.round(float) in full 32 bits range
+ *
+ * @library /test/lib /
+ * @modules java.base/jdk.internal.math
+ * @requires os.arch == "riscv64"
+ * @run main/othervm -XX:-TieredCompilation -Xbatch -XX:CompileThresholdScaling=0.3 -XX:-UseSuperWord
+ *      -XX:CompileCommand=compileonly,compiler.floatingpoint.TestRoundFloatAll::test*
+ *      compiler.floatingpoint.TestRoundFloatAll
+ */
+
+package compiler.floatingpoint;
+
+import static compiler.lib.golden.GoldenRound.golden_round;
+
+public class TestRoundFloatAll {
+
+  public static void main(String args[]) {
+    test();
+  }
+
+  // return true when test fails
+  static boolean test(int n, float f) {
+    int actual = Math.round(f);
+    int expected = golden_round(f);
+    if (actual != expected) {
+      System.err.println("round error, input: " + f + ", res: " + actual + "expected: " + expected + ", input hex: " + n);
+      return true;
+    }
+    return false;
+  }
+
+  static void test() {
+    final int ITERS = 11000;
+    boolean fail = false;
+
+    // Warmup
+    System.out.println("Warmup");
+    for (int i=0; i<ITERS; i++) {
+      float f = Float.intBitsToFloat(i);
+      fail |= test(i, f);
+    }
+    if (fail) {
+      throw new RuntimeException("Warmup failed");
+    }
+
+    // Test and verify results
+    System.out.println("Verification");
+    int testInt = 0;
+    do {
+      float testFloat = Float.intBitsToFloat(testInt);
+      fail |= test(testInt, testFloat);
+    } while (++testInt != 0);
+    if (fail) {
+      throw new RuntimeException("Test failed");
+    }
+  }
+}
diff --git a/test/hotspot/jtreg/compiler/lib/golden/GoldenRound.java b/test/hotspot/jtreg/compiler/lib/golden/GoldenRound.java
new file mode 100644
index 0000000000000..38bceeafe26f3
--- /dev/null
+++ b/test/hotspot/jtreg/compiler/lib/golden/GoldenRound.java
@@ -0,0 +1,95 @@
+/*
+ * Copyright (c) 2024, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2024, Rivos Inc. All rights reserved.
+ * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
+ *
+ * This code is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 only, as
+ * published by the Free Software Foundation.
+ *
+ * This code is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+ * version 2 for more details (a copy is included in the LICENSE file that
+ * accompanied this code).
+ *
+ * You should have received a copy of the GNU General Public License version
+ * 2 along with this work; if not, write to the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
+ * or visit www.oracle.com if you need additional information or have any
+ * questions.
+ */
+
+package compiler.lib.golden;
+
+import jdk.internal.math.DoubleConsts;
+import jdk.internal.math.FloatConsts;
+
+public class GoldenRound {
+  public static int golden_round(float a) {
+    // below code is copied from java.base/share/classes/java/lang/Math.java
+    //  public static int round(float a) { ... }
+
+    int intBits = Float.floatToRawIntBits(a);
+    int biasedExp = (intBits & FloatConsts.EXP_BIT_MASK)
+            >> (FloatConsts.SIGNIFICAND_WIDTH - 1);
+    int shift = (FloatConsts.SIGNIFICAND_WIDTH - 2
+            + FloatConsts.EXP_BIAS) - biasedExp;
+    if ((shift & -32) == 0) { // shift >= 0 && shift < 32
+        // a is a finite number such that pow(2,-32) <= ulp(a) < 1
+        int r = ((intBits & FloatConsts.SIGNIF_BIT_MASK)
+                | (FloatConsts.SIGNIF_BIT_MASK + 1));
+        if (intBits < 0) {
+            r = -r;
+        }
+        // In the comments below each Java expression evaluates to the value
+        // the corresponding mathematical expression:
+        // (r) evaluates to a / ulp(a)
+        // (r >> shift) evaluates to floor(a * 2)
+        // ((r >> shift) + 1) evaluates to floor((a + 1/2) * 2)
+        // (((r >> shift) + 1) >> 1) evaluates to floor(a + 1/2)
+        return ((r >> shift) + 1) >> 1;
+    } else {
+        // a is either
+        // - a finite number with abs(a) < exp(2,FloatConsts.SIGNIFICAND_WIDTH-32) < 1/2
+        // - a finite number with ulp(a) >= 1 and hence a is a mathematical integer
+        // - an infinity or NaN
+        return (int) a;
+    }
+  }
+
+
+  public static long golden_round(double a) {
+    // below code is copied from java.base/share/classes/java/lang/Math.java
+    //  public static int round(double a) { ... }
+
+    long longBits = Double.doubleToRawLongBits(a);
+    long biasedExp = (longBits & DoubleConsts.EXP_BIT_MASK)
+            >> (DoubleConsts.SIGNIFICAND_WIDTH - 1);
+    long shift = (DoubleConsts.SIGNIFICAND_WIDTH - 2
+            + DoubleConsts.EXP_BIAS) - biasedExp;
+    if ((shift & -64) == 0) { // shift >= 0 && shift < 64
+        // a is a finite number such that pow(2,-64) <= ulp(a) < 1
+        long r = ((longBits & DoubleConsts.SIGNIF_BIT_MASK)
+                | (DoubleConsts.SIGNIF_BIT_MASK + 1));
+        if (longBits < 0) {
+            r = -r;
+        }
+        // In the comments below each Java expression evaluates to the value
+        // the corresponding mathematical expression:
+        // (r) evaluates to a / ulp(a)
+        // (r >> shift) evaluates to floor(a * 2)
+        // ((r >> shift) + 1) evaluates to floor((a + 1/2) * 2)
+        // (((r >> shift) + 1) >> 1) evaluates to floor(a + 1/2)
+        return ((r >> shift) + 1) >> 1;
+    } else {
+        // a is either
+        // - a finite number with abs(a) < exp(2,DoubleConsts.SIGNIFICAND_WIDTH-64) < 1/2
+        // - a finite number with ulp(a) >= 1 and hence a is a mathematical integer
+        // - an infinity or NaN
+        return (long) a;
+    }
+  }
+}
diff --git a/test/hotspot/jtreg/compiler/lib/ir_framework/IRNode.java b/test/hotspot/jtreg/compiler/lib/ir_framework/IRNode.java
index 162f384dc07b2..24fbe7e64afa0 100644
--- a/test/hotspot/jtreg/compiler/lib/ir_framework/IRNode.java
+++ b/test/hotspot/jtreg/compiler/lib/ir_framework/IRNode.java
@@ -276,25 +276,25 @@ public class IRNode {
 
     public static final String ALLOC = PREFIX + "ALLOC" + POSTFIX;
     static {
-        String optoRegex = "(.*precise .*\\R((.*(?i:mov|mv|xorl|nop|spill).*|\\s*)\\R)*.*(?i:call,static).*wrapper for: _new_instance_Java" + END;
+        String optoRegex = "(.*precise .*\\R((.*(?i:mov|mv|xorl|nop|spill).*|\\s*)\\R)*.*(?i:call,static).*wrapper for: C2 Runtime new_instance" + END;
         allocNodes(ALLOC, "Allocate", optoRegex);
     }
 
     public static final String ALLOC_OF = COMPOSITE_PREFIX + "ALLOC_OF" + POSTFIX;
     static {
-        String regex = "(.*precise .*" + IS_REPLACED + ":.*\\R((.*(?i:mov|mv|xorl|nop|spill).*|\\s*)\\R)*.*(?i:call,static).*wrapper for: _new_instance_Java" + END;
+        String regex = "(.*precise .*" + IS_REPLACED + ":.*\\R((.*(?i:mov|mv|xorl|nop|spill).*|\\s*)\\R)*.*(?i:call,static).*wrapper for: C2 Runtime new_instance" + END;
         optoOnly(ALLOC_OF, regex);
     }
 
     public static final String ALLOC_ARRAY = PREFIX + "ALLOC_ARRAY" + POSTFIX;
     static {
-        String optoRegex = "(.*precise \\[.*\\R((.*(?i:mov|mv|xor|nop|spill).*|\\s*|.*(LGHI|LI).*)\\R)*.*(?i:call,static).*wrapper for: _new_array_Java" + END;
+        String optoRegex = "(.*precise \\[.*\\R((.*(?i:mov|mv|xor|nop|spill).*|\\s*|.*(LGHI|LI).*)\\R)*.*(?i:call,static).*wrapper for: C2 Runtime new_array" + END;
         allocNodes(ALLOC_ARRAY, "AllocateArray", optoRegex);
     }
 
     public static final String ALLOC_ARRAY_OF = COMPOSITE_PREFIX + "ALLOC_ARRAY_OF" + POSTFIX;
     static {
-        String regex = "(.*precise \\[.*" + IS_REPLACED + ":.*\\R((.*(?i:mov|mv|xorl|nop|spill).*|\\s*|.*(LGHI|LI).*)\\R)*.*(?i:call,static).*wrapper for: _new_array_Java" + END;
+        String regex = "(.*precise \\[.*" + IS_REPLACED + ":.*\\R((.*(?i:mov|mv|xorl|nop|spill).*|\\s*|.*(LGHI|LI).*)\\R)*.*(?i:call,static).*wrapper for: C2 Runtime new_array" + END;
         optoOnly(ALLOC_ARRAY_OF, regex);
     }
 
diff --git a/test/hotspot/jtreg/compiler/vectorization/TestRoundVectRiscv64.java b/test/hotspot/jtreg/compiler/vectorization/TestRoundVectRiscv64.java
new file mode 100644
index 0000000000000..530befbe2bcb5
--- /dev/null
+++ b/test/hotspot/jtreg/compiler/vectorization/TestRoundVectRiscv64.java
@@ -0,0 +1,94 @@
+/*
+ * Copyright (c) 2024, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2024, Rivos Inc. All rights reserved.
+ * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
+ *
+ * This code is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 only, as
+ * published by the Free Software Foundation.
+ *
+ * This code is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+ * version 2 for more details (a copy is included in the LICENSE file that
+ * accompanied this code).
+ *
+ * You should have received a copy of the GNU General Public License version
+ * 2 along with this work; if not, write to the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
+ * or visit www.oracle.com if you need additional information or have any
+ * questions.
+ */
+
+/**
+ * @test
+ * @bug 1234567
+ * @summary Auto-vectorize Math.round API
+ * @requires vm.compiler2.enabled
+ * @requires os.arch == "riscv64"
+ * @requires vm.cpu.features ~= ".*rvv.*"
+ * @library /test/lib /
+ * @run driver compiler.vectorization.TestRoundVectRiscv64
+ */
+
+package compiler.vectorization;
+
+import compiler.lib.ir_framework.*;
+
+public class TestRoundVectRiscv64 {
+  private static final int ARRLEN = 1024;
+  private static final int ITERS  = 11000;
+
+  private static double [] dinp;
+  private static long   [] lout;
+  private static float  [] finp;
+  private static int    [] iout;
+
+  public static void main(String args[]) {
+      TestFramework.runWithFlags("-XX:-TieredCompilation",
+                                 "-XX:CompileThresholdScaling=0.3");
+      System.out.println("PASSED");
+  }
+
+  @Test
+  @IR(counts = {IRNode.ROUND_VD, "> 0"})
+  public void test_round_double(long[] lout, double[] dinp) {
+      for (int i = 0; i < lout.length; i+=1) {
+          lout[i] = Math.round(dinp[i]);
+      }
+  }
+
+  @Run(test = {"test_round_double"}, mode = RunMode.STANDALONE)
+  public void kernel_test_round_double() {
+      dinp = new double[ARRLEN];
+      lout = new long[ARRLEN];
+      for (int i = 0 ; i < ARRLEN; i++) {
+          dinp[i] = (double)i*1.4;
+      }
+      for (int i = 0; i < ITERS; i++) {
+          test_round_double(lout , dinp);
+      }
+  }
+
+  @Test
+  @IR(counts = {IRNode.ROUND_VF, "> 0"})
+  public void test_round_float(int[] iout, float[] finp) {
+      for (int i = 0; i < finp.length; i+=1) {
+          iout[i] = Math.round(finp[i]);
+      }
+  }
+
+  @Run(test = {"test_round_float"}, mode = RunMode.STANDALONE)
+  public void kernel_test_round() {
+      finp = new float[ARRLEN];
+      iout = new int[ARRLEN];
+      for (int i = 0 ; i < ARRLEN; i++) {
+          finp[i] = (float)i*1.4f;
+      }
+      for (int i = 0; i < ITERS; i++) {
+          test_round_float(iout , finp);
+      }
+  }
+}
diff --git a/test/hotspot/jtreg/compiler/vectorization/TestRoundVectorDoubleRandom.java b/test/hotspot/jtreg/compiler/vectorization/TestRoundVectorDoubleRandom.java
new file mode 100644
index 0000000000000..cd1a59c6d24e0
--- /dev/null
+++ b/test/hotspot/jtreg/compiler/vectorization/TestRoundVectorDoubleRandom.java
@@ -0,0 +1,229 @@
+/*
+ * Copyright (c) 2024, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2024, Rivos Inc. All rights reserved.
+ * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
+ *
+ * This code is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 only, as
+ * published by the Free Software Foundation.
+ *
+ * This code is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+ * version 2 for more details (a copy is included in the LICENSE file that
+ * accompanied this code).
+ *
+ * You should have received a copy of the GNU General Public License version
+ * 2 along with this work; if not, write to the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
+ * or visit www.oracle.com if you need additional information or have any
+ * questions.
+ */
+
+/**
+ * @test
+ * @key randomness
+ * @bug 8321011
+ * @summary Test vector intrinsic for Math.round(double) with random input in 64 bits range, verify IR at the same time.
+ *
+ * @library /test/lib /
+ * @modules java.base/jdk.internal.math
+ * @requires os.arch == "riscv64" & vm.cpu.features ~= ".*rvv.*"
+ * @run main compiler.vectorization.TestRoundVectorDoubleRandom
+ */
+
+package compiler.vectorization;
+
+import java.util.Random;
+import static compiler.lib.golden.GoldenRound.golden_round;
+import compiler.lib.ir_framework.IR;
+import compiler.lib.ir_framework.IRNode;
+import compiler.lib.ir_framework.Run;
+import compiler.lib.ir_framework.RunInfo;
+import compiler.lib.ir_framework.Test;
+import compiler.lib.ir_framework.TestFramework;
+import compiler.lib.ir_framework.Warmup;
+
+public class TestRoundVectorDoubleRandom {
+  private static final Random rand = new Random();
+
+  private static final int ITERS  = 11000;
+  private static final int ARRLEN = rand.nextInt(4096-997) + 997;
+  private static final double ADD_INIT = -7500.;
+
+  private static final double[] input = new double[ARRLEN];
+  private static final long [] res = new long[ARRLEN];
+
+  public static void main(String args[]) {
+    TestFramework.runWithFlags("-XX:-TieredCompilation", "-XX:CompileThresholdScaling=0.3");
+    TestFramework.runWithFlags("-XX:-TieredCompilation", "-XX:CompileThresholdScaling=0.3", "-XX:MaxVectorSize=8");
+    TestFramework.runWithFlags("-XX:-TieredCompilation", "-XX:CompileThresholdScaling=0.3", "-XX:MaxVectorSize=16");
+    TestFramework.runWithFlags("-XX:-TieredCompilation", "-XX:CompileThresholdScaling=0.3", "-XX:MaxVectorSize=32");
+  }
+
+  @Test
+  @IR(counts = {IRNode.ROUND_VD, "> 0"},
+      applyIf = {"MaxVectorSize", ">= 64"})
+  static void test_round(long[] a0, double[] a1) {
+    for (int i = 0; i < a0.length; i+=1) {
+      a0[i] = Math.round(a1[i]);
+    }
+  }
+
+  @Run(test = "test_round")
+  @Warmup(ITERS)
+  static void test_rounds(RunInfo runInfo) {
+    // Initialize
+    for (int i = 0; i < ARRLEN; i++) {
+      double val = ADD_INIT+(double)i;
+      input[i] = val;
+    }
+
+    test_round(res, input);
+    // skip test/verify when warming up
+    if (runInfo.isWarmUp()) {
+      return;
+    }
+
+    int errn = 0;
+    // a double precise float point is composed of 3 parts: sign/exponent/signicand
+    // exponent part of a float value
+    final int exponentShift = 52;
+    final int exponentWidth = 11;
+    final int exponentBound = 1 << exponentWidth;
+    // significant part of a float value
+    final int signicandWidth = exponentShift;
+    final long signicandBound = 1L << signicandWidth;
+    final int signicandNum = 256;
+
+    // prepare for data of significand part
+    long signicandValues[] = new long[signicandNum];
+    int signicandIdx = 0;
+    for (; signicandIdx < signicandWidth; signicandIdx++) {
+      signicandValues[signicandIdx] = 1L << signicandIdx;
+    }
+    for (; signicandIdx < signicandNum; signicandIdx++) {
+      signicandValues[signicandIdx] = rand.nextLong(signicandBound);
+    }
+    signicandValues[rand.nextInt(signicandNum)] = 0;
+
+    // generate input arrays for testing, then run tests & verify results
+
+    // generate input arrays by combining different parts
+    for (long sv : signicandValues) {
+      // generate test input by combining different parts:
+      //   previously generated significand values,
+      //   random value in exponent range,
+      //   both positive and negative of previous combined values (exponent+significand)
+      final int exponentStart = rand.nextInt(9);
+      final int exponentStep = (1 << 3) + rand.nextInt(3);
+      // Here, we could have iterated the whole range of exponent values, but it would
+      // take more time to run the test, so just randomly choose some of exponent values.
+      int ev = exponentStart;
+      int inputIdx = 0;
+      for (; ev < exponentBound; ev += exponentStep) {
+        inputIdx = ev/exponentStep;
+        // combine exponent and significand
+        long bits = ((long)ev << exponentShift) + sv;
+        // combine sign(+/-) with exponent and significand
+        // positive values
+        input[inputIdx*2] = Double.longBitsToDouble(bits);
+        // negative values
+        bits = bits | (1L << 63);
+        input[inputIdx*2+1] = Double.longBitsToDouble(bits);
+      }
+      // add specific test cases where it looks like in binary format:
+      //   s111 1111 1111 xxxx xxxx xxxx xxxx xxxx ...
+      // these are for the NaN and Inf.
+      inputIdx = inputIdx*2+2;
+      long bits = (1L << exponentWidth) - 1L;
+      bits <<= exponentShift;
+      input[inputIdx++] = Double.longBitsToDouble(bits);
+      bits = bits | (1L << 63);
+      input[inputIdx] = Double.longBitsToDouble(bits);
+
+      // run tests
+      test_round(res, input);
+
+      // verify results
+      ev = exponentStart;
+      inputIdx = ev/exponentStep;
+      for (; ev < exponentBound; ev += exponentStep) {
+        for (int sign = 0; sign < 2; sign++) {
+          int idx = inputIdx * 2 + sign;
+          if (res[idx] != golden_round(input[idx])) {
+            errn++;
+            System.err.println("round error, input: " + input[idx] +
+                               ", res: " + res[idx] + "expected: " + golden_round(input[idx]) +
+                               ", input hex: " + Double.doubleToLongBits(input[idx]) +
+                               ", fi: " + sv + ", ei: " + ev + ", sign: " + sign);
+          }
+        }
+      }
+    }
+
+    // generate pure random input arrays, which does not depend on significand/exponent values
+    for(int i = 0; i < 128; i++) {
+      for (int j = 0; j < ARRLEN; j++) {
+        input[j] = rand.nextDouble();
+      }
+
+      // run tests
+      test_round(res, input);
+
+      // verify results
+      for (int j = 0; j < ARRLEN; j++) {
+        if (res[j] != golden_round(input[j])) {
+          errn++;
+          System.err.println("round error, input: " + input[j] +
+                             ", res: " + res[j] + "expected: " + golden_round(input[j]) +
+                             ", input hex: " + Double.doubleToLongBits(input[j]));
+        }
+      }
+    }
+
+    // test cases for NaN, Inf, subnormal, and so on
+    {
+      Double[] dv = new Double[] {
+        +0.0,
+        -0.0,
+        Double.MAX_VALUE,
+        Double.MIN_VALUE,
+        Double.NEGATIVE_INFINITY,
+        Double.POSITIVE_INFINITY,
+        Double.NaN,
+        Double.longBitsToDouble(0x7ff0000000000001L), // another NaN
+        Double.MIN_NORMAL,
+        0x0.fffffffffffffp-1022,   // Maximum Subnormal Value
+        1.5,
+        100.5,
+        10000.5,
+        -1.5,
+        -100.5,
+        -10000.5
+      };
+      for (int j = 0; j < ARRLEN; j++) {
+        input[j] = dv[rand.nextInt(dv.length)];
+      }
+
+      // run tests
+      test_round(res, input);
+
+      // verify results
+      for (int j = 0; j < ARRLEN; j++) {
+        if (res[j] != golden_round(input[j])) {
+          errn++;
+          System.err.println("round error, input: " + input[j] +
+                             ", res: " + res[j] + "expected: " + golden_round(input[j]) +
+                             ", input hex: " + Double.doubleToLongBits(input[j]));
+        }
+      }
+    }
+
+    if (errn > 0) {
+      throw new RuntimeException("There are some round error detected!");
+    }
+  }
+}
diff --git a/test/hotspot/jtreg/compiler/vectorization/TestRoundVectorFloatAll.java b/test/hotspot/jtreg/compiler/vectorization/TestRoundVectorFloatAll.java
new file mode 100644
index 0000000000000..a85ac3eaa22e0
--- /dev/null
+++ b/test/hotspot/jtreg/compiler/vectorization/TestRoundVectorFloatAll.java
@@ -0,0 +1,95 @@
+/*
+ * Copyright (c) 2024, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2024, Rivos Inc. All rights reserved.
+ * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
+ *
+ * This code is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 only, as
+ * published by the Free Software Foundation.
+ *
+ * This code is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+ * version 2 for more details (a copy is included in the LICENSE file that
+ * accompanied this code).
+ *
+ * You should have received a copy of the GNU General Public License version
+ * 2 along with this work; if not, write to the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
+ * or visit www.oracle.com if you need additional information or have any
+ * questions.
+ */
+
+/**
+ * @test
+ * @bug 8321010
+ * @summary Test vector intrinsic for Math.round(float) in full 32 bits range
+ *
+ * @library /test/lib /
+ * @modules java.base/jdk.internal.math
+ * @requires os.arch == "riscv64" & vm.cpu.features ~= ".*rvv.*"
+ * @run main/othervm -XX:-TieredCompilation -XX:CompileThresholdScaling=0.3 -XX:+UseSuperWord -XX:CompileCommand=compileonly,compiler.vectorization.TestRoundVectorFloatAll::test* compiler.vectorization.TestRoundVectorFloatAll
+ * @run main/othervm -XX:-TieredCompilation -XX:CompileThresholdScaling=0.3 -XX:MaxVectorSize=32 -XX:+UseSuperWord -XX:CompileCommand=compileonly,compiler.vectorization.TestRoundVectorFloatAll::test* compiler.vectorization.TestRoundVectorFloatAll
+ */
+
+package compiler.vectorization;
+
+import java.util.Random;
+import static compiler.lib.golden.GoldenRound.golden_round;
+
+public class TestRoundVectorFloatAll {
+  private static final Random rand = new Random();
+
+  private static final int ITERS  = 11000;
+  private static final int ARRLEN = rand.nextInt(4096-997) + 997;
+  private static final float ADD_INIT = -7500.f;
+
+  public static void main(String args[]) {
+    test();
+  }
+
+  static void test() {
+    float[] input = new float[ARRLEN];
+    int [] res = new int[ARRLEN];
+
+    // Warmup
+    System.out.println("Warmup");
+    for (int i=0; i<ITERS; i++) {
+      test_round(res, input);
+    }
+
+    // Test and verify results
+    System.out.println("Verification");
+    int errn = 0;
+    for (long l = Integer.MIN_VALUE; l <= Integer.MAX_VALUE; l+=ARRLEN) {
+      for (int i = 0; i < ARRLEN; i++) {
+        input[i] = Float.intBitsToFloat((int)(i+l));
+      }
+
+      // run tests
+      test_round(res, input);
+
+      // verify results
+      for (int j = 0; j < ARRLEN; j++) {
+        if (res[j] != golden_round(input[j])) {
+          errn++;
+          System.err.println("round error, input: " + input[j] +
+                             ", res: " + res[j] + "expected: " + golden_round(input[j]) +
+                             ", input hex: " + Float.floatToIntBits(input[j]));
+        }
+      }
+    }
+
+    if (errn > 0) {
+      throw new RuntimeException("There are some round error detected!");
+    }
+  }
+
+  static void test_round(int[] a0, float[] a1) {
+    for (int i = 0; i < a0.length; i+=1) {
+      a0[i] = Math.round(a1[i]);
+    }
+  }
+}
diff --git a/test/hotspot/jtreg/compiler/vectorization/TestRoundVectorFloatRandom.java b/test/hotspot/jtreg/compiler/vectorization/TestRoundVectorFloatRandom.java
new file mode 100644
index 0000000000000..7288e4d9f6396
--- /dev/null
+++ b/test/hotspot/jtreg/compiler/vectorization/TestRoundVectorFloatRandom.java
@@ -0,0 +1,212 @@
+/*
+ * Copyright (c) 2024, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2024, Rivos Inc. All rights reserved.
+ * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
+ *
+ * This code is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 only, as
+ * published by the Free Software Foundation.
+ *
+ * This code is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+ * version 2 for more details (a copy is included in the LICENSE file that
+ * accompanied this code).
+ *
+ * You should have received a copy of the GNU General Public License version
+ * 2 along with this work; if not, write to the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
+ * or visit www.oracle.com if you need additional information or have any
+ * questions.
+ */
+
+/**
+ * @test
+ * @key randomness
+ * @bug 8321010
+ * @summary Test vector intrinsic for Math.round(float) with random input in 32 bits range, verify IR at the same time.
+ *
+ * @library /test/lib /
+ * @modules java.base/jdk.internal.math
+ * @requires os.arch == "riscv64" & vm.cpu.features ~= ".*rvv.*"
+ * @run main compiler.vectorization.TestRoundVectorFloatRandom
+ */
+
+package compiler.vectorization;
+
+import java.util.Random;
+import static compiler.lib.golden.GoldenRound.golden_round;
+import compiler.lib.ir_framework.IR;
+import compiler.lib.ir_framework.IRNode;
+import compiler.lib.ir_framework.Run;
+import compiler.lib.ir_framework.RunInfo;
+import compiler.lib.ir_framework.Test;
+import compiler.lib.ir_framework.TestFramework;
+import compiler.lib.ir_framework.Warmup;
+
+public class TestRoundVectorFloatRandom {
+  private static final Random rand = new Random();
+
+  private static final int ITERS  = 11000;
+  private static final int ARRLEN = rand.nextInt(4096-997) + 997;
+  private static final float ADD_INIT = -7500.f;
+
+  private static final float[] input = new float[ARRLEN];
+  private static final int[] res = new int[ARRLEN];
+
+  public static void main(String args[]) {
+    TestFramework.runWithFlags("-XX:-TieredCompilation", "-XX:CompileThresholdScaling=0.3");
+    TestFramework.runWithFlags("-XX:-TieredCompilation", "-XX:CompileThresholdScaling=0.3", "-XX:MaxVectorSize=8");
+    TestFramework.runWithFlags("-XX:-TieredCompilation", "-XX:CompileThresholdScaling=0.3", "-XX:MaxVectorSize=16");
+    TestFramework.runWithFlags("-XX:-TieredCompilation", "-XX:CompileThresholdScaling=0.3", "-XX:MaxVectorSize=32");
+  }
+
+  @Test
+  @IR(counts = {IRNode.ROUND_VF, "> 0"},
+      applyIf = {"MaxVectorSize", ">= 32"})
+  static void test_round(int[] a0, float[] a1) {
+    for (int i = 0; i < a0.length; i+=1) {
+      a0[i] = Math.round(a1[i]);
+    }
+  }
+
+  @Run(test = "test_round")
+  @Warmup(ITERS)
+  static void test_rounds(RunInfo runInfo) {
+    // Initialize
+    for (int i = 0; i < ARRLEN; i++) {
+      float val = ADD_INIT+(float)i;
+      input[i] = val;
+    }
+
+    test_round(res, input);
+    // skip test/verify when warming up
+    if (runInfo.isWarmUp()) {
+      return;
+    }
+
+    int errn = 0;
+    // a single precise float point is composed of 3 parts: sign/exponent/signicand
+    // exponent part of a float value
+    final int exponentStart = 0;
+    final int exponentShift = 23;
+    final int exponentWidth = 8;
+    final int exponentBound = 1 << exponentWidth;
+    // significant part of a float value
+    final int signicandWidth = exponentShift;
+    final int signicandBound = 1 << signicandWidth;
+    final int signicandNum = 128;
+
+    // prepare for data of significand part
+    int signicandValues[] = new int[signicandNum];
+    int signicandIdx = 0;
+    for (; signicandIdx < signicandWidth; signicandIdx++) {
+      signicandValues[signicandIdx] = 1 << signicandIdx;
+    }
+    for (; signicandIdx < signicandNum; signicandIdx++) {
+      signicandValues[signicandIdx] = rand.nextInt(signicandBound);
+    }
+    signicandValues[rand.nextInt(signicandNum)] = 0;
+
+    // generate input arrays for testing, then run tests & verify results
+
+    // generate input arrays by combining different parts
+    for (int sv : signicandValues) {
+      // generate test input by combining different parts:
+      //   previously generated significand values,
+      //   all values in exponent range,
+      //   both positive and negative of previous combined values (exponent+significand)
+      for (int ev = exponentStart; ev < exponentBound; ev++) {
+        // combine exponent and significand
+        int bits = (ev << exponentShift) + sv;
+        // combine sign(+/-) with exponent and significand
+        // positive values
+        input[ev*2] = Float.intBitsToFloat(bits);
+        // negative values
+        bits = bits | (1 << 31);
+        input[ev*2+1] = Float.intBitsToFloat(bits);
+      }
+
+      // run tests
+      test_round(res, input);
+
+      // verify results
+      for (int ev = exponentStart; ev < exponentBound; ev++) {
+        for (int sign = 0; sign < 2; sign++) {
+          int idx = ev * 2 + sign;
+          if (res[idx] != golden_round(input[idx])) {
+            errn++;
+            System.err.println("round error, input: " + input[idx] +
+                               ", res: " + res[idx] + "expected: " + golden_round(input[idx]) +
+                               ", input hex: " + Float.floatToIntBits(input[idx]) +
+                               ", fi: " + sv + ", ei: " + ev + ", sign: " + sign);
+          }
+        }
+      }
+    }
+
+    // generate pure random input arrays, which does not depend on significand/exponent values
+    for(int i = 0; i < 128; i++) {
+      for (int j = 0; j < ARRLEN; j++) {
+        input[j] = rand.nextFloat();
+      }
+
+      // run tests
+      test_round(res, input);
+
+      // verify results
+      for (int j = 0; j < ARRLEN; j++) {
+        if (res[j] != golden_round(input[j])) {
+          errn++;
+          System.err.println("round error, input: " + input[j] +
+                             ", res: " + res[j] + "expected: " + golden_round(input[j]) +
+                             ", input hex: " + Float.floatToIntBits(input[j]));
+        }
+      }
+    }
+
+    // test cases for NaN, Inf, subnormal, and so on
+    {
+      Float[] dv = new Float[] {
+        +0.0f,
+        -0.0f,
+        Float.MAX_VALUE,
+        Float.MIN_VALUE,
+        Float.NEGATIVE_INFINITY,
+        Float.POSITIVE_INFINITY,
+        Float.NaN,
+        Float.intBitsToFloat(0x7f800001), // another NaN
+        Float.MIN_NORMAL,
+        0x0.fffffep-126f,   // Maximum Subnormal Value
+        1.5f,
+        100.5f,
+        10000.5f,
+        -1.5f,
+        -100.5f,
+        -10000.5f
+      };
+      for (int j = 0; j < ARRLEN; j++) {
+        input[j] = dv[rand.nextInt(dv.length)];
+      }
+
+      // run tests
+      test_round(res, input);
+
+      // verify results
+      for (int j = 0; j < ARRLEN; j++) {
+        if (res[j] != golden_round(input[j])) {
+          errn++;
+          System.err.println("round error, input: " + input[j] +
+                             ", res: " + res[j] + "expected: " + golden_round(input[j]) +
+                             ", input hex: " + Float.floatToIntBits(input[j]));
+        }
+      }
+    }
+
+    if (errn > 0) {
+      throw new RuntimeException("There are some round error detected!");
+    }
+  }
+}
diff --git a/test/hotspot/jtreg/gc/x/TestAllocateHeapAt.java b/test/hotspot/jtreg/gc/x/TestAllocateHeapAt.java
index 3bf83d9076864..bacd7a2078e97 100644
--- a/test/hotspot/jtreg/gc/x/TestAllocateHeapAt.java
+++ b/test/hotspot/jtreg/gc/x/TestAllocateHeapAt.java
@@ -41,7 +41,7 @@ public static void main(String[] args) throws Exception {
         final String heapBackingFile = "Heap Backing File: " + directory;
         final String failedToCreateFile = "Failed to create file " + directory;
 
-        ProcessTools.executeLimitedTestJava(
+        ProcessTools.executeTestJava(
             "-XX:+UseZGC",
             "-XX:-ZGenerational",
             "-Xlog:gc*",
diff --git a/test/hotspot/jtreg/gc/x/TestPageCacheFlush.java b/test/hotspot/jtreg/gc/x/TestPageCacheFlush.java
index cb8685d0d090c..a48b6f77e17ef 100644
--- a/test/hotspot/jtreg/gc/x/TestPageCacheFlush.java
+++ b/test/hotspot/jtreg/gc/x/TestPageCacheFlush.java
@@ -68,7 +68,7 @@ public static void main(String[] args) throws Exception {
     }
 
     public static void main(String[] args) throws Exception {
-        ProcessTools.executeLimitedTestJava(
+        ProcessTools.executeTestJava(
             "-XX:+UseZGC",
             "-XX:-ZGenerational",
             "-Xms128M",
diff --git a/test/hotspot/jtreg/gc/x/TestSmallHeap.java b/test/hotspot/jtreg/gc/x/TestSmallHeap.java
index 8fc6f07be398f..a7e8042f92474 100644
--- a/test/hotspot/jtreg/gc/x/TestSmallHeap.java
+++ b/test/hotspot/jtreg/gc/x/TestSmallHeap.java
@@ -53,7 +53,7 @@ public static void main(String[] args) throws Exception {
 
     public static void main(String[] args) throws Exception {
         for (var maxCapacity: args) {
-            ProcessTools.executeLimitedTestJava(
+            ProcessTools.executeTestJava(
                 "-XX:+UseZGC",
                 "-XX:-ZGenerational",
                 "-Xlog:gc,gc+init,gc+reloc,gc+heap",
diff --git a/test/hotspot/jtreg/gc/z/TestAllocateHeapAt.java b/test/hotspot/jtreg/gc/z/TestAllocateHeapAt.java
index 9f47c4b60d3cb..28d2ebd6aea01 100644
--- a/test/hotspot/jtreg/gc/z/TestAllocateHeapAt.java
+++ b/test/hotspot/jtreg/gc/z/TestAllocateHeapAt.java
@@ -41,7 +41,7 @@ public static void main(String[] args) throws Exception {
         final String heapBackingFile = "Heap Backing File: " + directory;
         final String failedToCreateFile = "Failed to create file " + directory;
 
-        ProcessTools.executeLimitedTestJava(
+        ProcessTools.executeTestJava(
             "-XX:+UseZGC",
             "-XX:+ZGenerational",
             "-Xlog:gc*",
diff --git a/test/hotspot/jtreg/gc/z/TestPageCacheFlush.java b/test/hotspot/jtreg/gc/z/TestPageCacheFlush.java
index 387053b580ad8..3b666ddc2c8a6 100644
--- a/test/hotspot/jtreg/gc/z/TestPageCacheFlush.java
+++ b/test/hotspot/jtreg/gc/z/TestPageCacheFlush.java
@@ -68,7 +68,7 @@ public static void main(String[] args) throws Exception {
     }
 
     public static void main(String[] args) throws Exception {
-        ProcessTools.executeLimitedTestJava(
+        ProcessTools.executeTestJava(
             "-XX:+UseZGC",
             "-XX:+ZGenerational",
             "-Xms128M",
diff --git a/test/hotspot/jtreg/gc/z/TestSmallHeap.java b/test/hotspot/jtreg/gc/z/TestSmallHeap.java
index 354cd4164f1fd..67d9d33d2815e 100644
--- a/test/hotspot/jtreg/gc/z/TestSmallHeap.java
+++ b/test/hotspot/jtreg/gc/z/TestSmallHeap.java
@@ -53,7 +53,7 @@ public static void main(String[] args) throws Exception {
 
     public static void main(String[] args) throws Exception {
         for (var maxCapacity: args) {
-            ProcessTools.executeLimitedTestJava(
+            ProcessTools.executeTestJava(
                 "-XX:+UseZGC",
                 "-XX:+ZGenerational",
                 "-Xlog:gc,gc+init,gc+reloc,gc+heap",
diff --git a/test/hotspot/jtreg/testlibrary_tests/ir_framework/tests/TestIRMatching.java b/test/hotspot/jtreg/testlibrary_tests/ir_framework/tests/TestIRMatching.java
index f77be989421b2..f45f708070ab0 100644
--- a/test/hotspot/jtreg/testlibrary_tests/ir_framework/tests/TestIRMatching.java
+++ b/test/hotspot/jtreg/testlibrary_tests/ir_framework/tests/TestIRMatching.java
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2021, 2023, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2021, 2024, Oracle and/or its affiliates. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -75,7 +75,7 @@ public static void main(String[] args) {
         runCheck(new String[] {"-XX:TLABRefillWasteFraction=50", "-XX:+UsePerfData", "-XX:+UseTLAB"}, BadFailOnConstraint.create(AndOr1.class, "test1(int)", 1, "CallStaticJava"));
         runCheck(new String[] {"-XX:TLABRefillWasteFraction=50", "-XX:-UsePerfData", "-XX:+UseTLAB"}, BadFailOnConstraint.create(AndOr1.class, "test2()", 1, "CallStaticJava"));
 
-        String[] allocMatches = { "MyClass", "wrapper for: _new_instance_Java" };
+        String[] allocMatches = { "MyClass", "wrapper for: C2 Runtime new_instance" };
         runCheck(BadFailOnConstraint.create(MultipleFailOnBad.class, "fail1()", 1, 1, "Store"),
                  BadFailOnConstraint.create(MultipleFailOnBad.class, "fail1()", 1,  3, "Store"),
                  GoodFailOnRegexConstraint.create(MultipleFailOnBad.class, "fail1()", 1,  2, 4),
@@ -114,7 +114,7 @@ public static void main(String[] args) {
                  GoodRuleConstraint.create(Calls.class, "calls()", 3)
         );
 
-        String[] allocArrayMatches = { "MyClass", "wrapper for: _new_array_Java"};
+        String[] allocArrayMatches = { "MyClass", "wrapper for: C2 Runtime new_array"};
         runCheck(BadFailOnConstraint.create(AllocArray.class, "allocArray()", 1, allocArrayMatches),
                  BadFailOnConstraint.create(AllocArray.class, "allocArray()", 2,  allocArrayMatches),
                  GoodFailOnConstraint.create(AllocArray.class, "allocArray()", 3),
diff --git a/test/hotspot/jtreg/vmTestbase/nsk/monitoring/share/ThreadController.java b/test/hotspot/jtreg/vmTestbase/nsk/monitoring/share/ThreadController.java
index ea15f3d0d35bc..9d121b9a94f79 100644
--- a/test/hotspot/jtreg/vmTestbase/nsk/monitoring/share/ThreadController.java
+++ b/test/hotspot/jtreg/vmTestbase/nsk/monitoring/share/ThreadController.java
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2003, 2023, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2003, 2024, Oracle and/or its affiliates. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -663,6 +663,9 @@ public SleepingThread(ThreadController controller, String name, Log log, Threads
         expectedMethods.add(Thread.class.getName() + ".currentCarrierThread");
         expectedMethods.add(Thread.class.getName() + ".currentThread");
         // jdk.internal.event.ThreadSleepEvent not accessible
+        expectedMethods.add("java.lang.Object.<init>");
+        expectedMethods.add("jdk.internal.event.Event.<init>");
+        expectedMethods.add("jdk.internal.event.ThreadSleepEvent.<init>");
         expectedMethods.add("jdk.internal.event.ThreadSleepEvent.<clinit>");
         expectedMethods.add("jdk.internal.event.ThreadSleepEvent.isEnabled");
         expectedMethods.add(SleepingThread.class.getName() + ".run");
diff --git a/test/hotspot/jtreg/vmTestbase/nsk/monitoring/share/thread/SleepingThread.java b/test/hotspot/jtreg/vmTestbase/nsk/monitoring/share/thread/SleepingThread.java
index ed2a1a60c5088..217c2cdfdc82b 100644
--- a/test/hotspot/jtreg/vmTestbase/nsk/monitoring/share/thread/SleepingThread.java
+++ b/test/hotspot/jtreg/vmTestbase/nsk/monitoring/share/thread/SleepingThread.java
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2007, 2023, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2007, 2024, Oracle and/or its affiliates. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -41,6 +41,9 @@ public class SleepingThread extends RecursiveMonitoringThread {
                 "java.lang.Thread.beforeSleep",
                 "java.lang.Thread.afterSleep",
                 "java.util.concurrent.TimeUnit.toNanos",
+                "java.lang.Object.<init>",
+                "jdk.internal.event.Event.<init>",
+                "jdk.internal.event.ThreadSleepEvent.<init>",
                 "jdk.internal.event.ThreadSleepEvent.<clinit>",
                 "jdk.internal.event.ThreadSleepEvent.isEnabled",
                 "nsk.monitoring.share.thread.SleepingThread.runInside"
diff --git a/test/hotspot/jtreg/vmTestbase/nsk/monitoring/stress/thread/strace001.java b/test/hotspot/jtreg/vmTestbase/nsk/monitoring/stress/thread/strace001.java
index 03d2066cf3289..0700032f6a7c2 100644
--- a/test/hotspot/jtreg/vmTestbase/nsk/monitoring/stress/thread/strace001.java
+++ b/test/hotspot/jtreg/vmTestbase/nsk/monitoring/stress/thread/strace001.java
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2003, 2023, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2003, 2024, Oracle and/or its affiliates. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -150,6 +150,8 @@ private static boolean fillTrace() {
                 "java.lang.Thread.currentThread",
                 "java.util.concurrent.TimeUnit.toNanos",
                 "jdk.internal.event.ThreadSleepEvent.<clinit>",
+                "java.lang.Object.<init>",
+                "jdk.internal.event.Event.<init>",
                 "jdk.internal.event.ThreadSleepEvent.<init>",
                 "jdk.internal.event.ThreadSleepEvent.isEnabled"
         };
diff --git a/test/jdk/ProblemList.txt b/test/jdk/ProblemList.txt
index 47a2e13b460cf..58949cf728256 100644
--- a/test/jdk/ProblemList.txt
+++ b/test/jdk/ProblemList.txt
@@ -133,6 +133,9 @@ java/awt/Focus/MouseClickRequestFocusRaceTest/MouseClickRequestFocusRaceTest.jav
 java/awt/Focus/NoAutotransferToDisabledCompTest/NoAutotransferToDisabledCompTest.java 7152980 macosx-all
 java/awt/Focus/ToFrontFocusTest/ToFrontFocus.java 7156130 linux-all
 java/awt/Focus/WrongKeyTypedConsumedTest/WrongKeyTypedConsumedTest.java 8169096 macosx-all
+java/awt/Focus/TestDisabledAutoTransfer.java 8159871 macosx-all,windows-all
+java/awt/Focus/TestDisabledAutoTransferSwing.java 6962362 windows-all
+java/awt/Focus/ActivateOnProperAppContextTest.java 8136516 macosx-all
 java/awt/EventQueue/6980209/bug6980209.java 8198615 macosx-all
 java/awt/grab/EmbeddedFrameTest1/EmbeddedFrameTest1.java 7080150 macosx-all
 java/awt/event/InputEvent/EventWhenTest/EventWhenTest.java 8168646 generic-all
diff --git a/test/jdk/java/awt/Choice/ChoiceInsertTest.java b/test/jdk/java/awt/Choice/ChoiceInsertTest.java
new file mode 100644
index 0000000000000..5eafa83a13991
--- /dev/null
+++ b/test/jdk/java/awt/Choice/ChoiceInsertTest.java
@@ -0,0 +1,89 @@
+/*
+ * Copyright (c) 1998, 2024, Oracle and/or its affiliates. All rights reserved.
+ * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
+ *
+ * This code is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 only, as
+ * published by the Free Software Foundation.
+ *
+ * This code is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+ * version 2 for more details (a copy is included in the LICENSE file that
+ * accompanied this code).
+ *
+ * You should have received a copy of the GNU General Public License version
+ * 2 along with this work; if not, write to the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
+ * or visit www.oracle.com if you need additional information or have any
+ * questions.
+ */
+
+import java.awt.Choice;
+import java.awt.EventQueue;
+import java.awt.Frame;
+import java.awt.GridLayout;
+import java.awt.Label;
+import java.awt.Robot;
+
+/*
+ * @test
+ * @bug 4082078
+ * @summary Test for bug(s): 4082078, Multiple calls to Choice.insert cause core dump
+ * @key headful
+ * @run main ChoiceInsertTest
+ */
+
+public class ChoiceInsertTest extends Frame {
+    Choice c;
+    Label  l;
+
+    private static ChoiceInsertTest choiceInsertTest;
+
+    public ChoiceInsertTest() {
+        c = new Choice();
+        l = new Label("If you see this, the choice insert bug is fixed!");
+        c.add("Initial choice");
+        add(c);
+    }
+
+    public void testInsertion() {
+        // inserting 30 or so items aborts Solaris VM
+        // in JDK's before 1.1.5
+        for (int nchoice = 0; nchoice < 30; nchoice++) {
+            c.insert("new choice", 0);
+        }
+        // if you made it to here the bug is not there anymore...
+        remove(l);
+        add(l);
+        validate();
+    }
+
+    public static void main(String[] args) throws Exception {
+        Robot robot = new Robot();
+        try {
+            EventQueue.invokeAndWait(() ->{
+                choiceInsertTest = new ChoiceInsertTest();
+                choiceInsertTest.setTitle("ChoiceInsertTest");
+                choiceInsertTest.setLocationRelativeTo(null);
+                choiceInsertTest.setSize(500, 300);
+                choiceInsertTest.setLayout(new GridLayout());
+                choiceInsertTest.setVisible(true);
+            });
+            robot.waitForIdle();
+            robot.delay(500);
+            EventQueue.invokeAndWait(choiceInsertTest::testInsertion);
+            robot.delay(1000);
+        } finally {
+            EventQueue.invokeAndWait(() -> {
+                if (choiceInsertTest != null) {
+                    choiceInsertTest.dispose();
+                }
+            });
+        }
+
+        System.err.println("ChoiceInsertTest: Didn't abort VM inserting 30 items, so we passed!");
+    }
+}
diff --git a/test/jdk/java/awt/Choice/ChoiceMouseDragTest.java b/test/jdk/java/awt/Choice/ChoiceMouseDragTest.java
new file mode 100644
index 0000000000000..0c0c059032dda
--- /dev/null
+++ b/test/jdk/java/awt/Choice/ChoiceMouseDragTest.java
@@ -0,0 +1,138 @@
+/*
+ * Copyright (c) 2000, 2024, Oracle and/or its affiliates. All rights reserved.
+ * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
+ *
+ * This code is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 only, as
+ * published by the Free Software Foundation.
+ *
+ * This code is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+ * version 2 for more details (a copy is included in the LICENSE file that
+ * accompanied this code).
+ *
+ * You should have received a copy of the GNU General Public License version
+ * 2 along with this work; if not, write to the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
+ * or visit www.oracle.com if you need additional information or have any
+ * questions.
+ */
+
+
+import java.awt.BorderLayout;
+import java.awt.Choice;
+import java.awt.EventQueue;
+import java.awt.Frame;
+import java.awt.Point;
+import java.awt.event.InputEvent;
+import java.awt.event.MouseAdapter;
+import java.awt.event.MouseEvent;
+
+/*
+ * @test
+ * @bug 4328557
+ * @summary Tests that MouseDragged and MouseReleased are triggered on choice
+ * @library /lib/client
+ * @build ExtendedRobot
+ * @key headful
+ * @run main ChoiceMouseDragTest
+ */
+
+
+public class ChoiceMouseDragTest extends Frame {
+    private static final Choice choice = new Choice();
+
+    private static ExtendedRobot robot;
+    private volatile boolean isDragged;
+    private volatile boolean isReleased;
+
+    private static volatile ChoiceMouseDragTest choiceMouseDragTest;
+
+    public ChoiceMouseDragTest() {
+        super("ChoiceMouseDragTest");
+        this.setLayout(new BorderLayout());
+        choice.add("item-1");
+        choice.add("item-2");
+        choice.add("item-3");
+        choice.add("item-4");
+        add("Center", choice);
+        choice.addMouseListener(new MouseEventHandler());
+        choice.addMouseMotionListener(new MouseMotionEventHandler());
+        setSize(400, 200);
+        setLocationRelativeTo(null);
+        setVisible(true);
+    }
+
+    public static void main(String[] args) throws Exception {
+        try {
+            EventQueue.invokeAndWait(() ->
+                    choiceMouseDragTest = new ChoiceMouseDragTest());
+
+            robot = new ExtendedRobot();
+            robot.waitForIdle();
+            robot.delay(500);
+
+            Point pointToDrag = choice.getLocationOnScreen();
+            pointToDrag.x += choice.getWidth() - 10;
+            pointToDrag.y += choice.getHeight() / 2 ;
+
+            choiceMouseDragTest.test(InputEvent.BUTTON3_DOWN_MASK, pointToDrag);
+            choiceMouseDragTest.test(InputEvent.BUTTON1_DOWN_MASK, pointToDrag);
+        } finally {
+            EventQueue.invokeAndWait(() -> {
+                if (choiceMouseDragTest != null) {
+                    choiceMouseDragTest.dispose();
+                }
+            });
+        }
+    }
+
+    void test(int buttonToTest, Point pointToDrag) {
+        isDragged = false;
+        isReleased = false;
+
+        robot.mouseMove(pointToDrag.x, pointToDrag.y);
+        robot.waitForIdle();
+
+        robot.mousePress(buttonToTest);
+
+        robot.glide(pointToDrag.x + 100, pointToDrag.y);
+        robot.waitForIdle();
+
+        robot.mouseRelease(buttonToTest);
+        robot.waitForIdle();
+
+        if (!isReleased || !isDragged) {
+            throw new RuntimeException(("Test failed: button %d dragged(received %b) or " +
+                    "released(received %b)")
+                    .formatted(buttonToTest, isDragged, isReleased));
+        }
+
+        robot.delay(500);
+    }
+
+    class MouseEventHandler extends MouseAdapter {
+        public void mousePressed(MouseEvent me) {
+            System.out.println(me.paramString());
+        }
+
+        public void mouseReleased(MouseEvent me) {
+            System.out.println(me.paramString());
+            isReleased = true;
+        }
+
+        public void mouseClicked(MouseEvent me) {
+            System.out.println(me.paramString());
+        }
+    }
+
+    class MouseMotionEventHandler extends MouseAdapter {
+        public void mouseDragged(MouseEvent me) {
+            System.out.println(me.paramString());
+            isDragged = true;
+        }
+    }
+}
diff --git a/test/jdk/java/awt/Choice/WheelEventsConsumed.java b/test/jdk/java/awt/Choice/WheelEventsConsumed.java
new file mode 100644
index 0000000000000..37200214734bb
--- /dev/null
+++ b/test/jdk/java/awt/Choice/WheelEventsConsumed.java
@@ -0,0 +1,156 @@
+/*
+ * Copyright (c) 2005, 2024, Oracle and/or its affiliates. All rights reserved.
+ * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
+ *
+ * This code is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 only, as
+ * published by the Free Software Foundation.
+ *
+ * This code is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+ * version 2 for more details (a copy is included in the LICENSE file that
+ * accompanied this code).
+ *
+ * You should have received a copy of the GNU General Public License version
+ * 2 along with this work; if not, write to the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
+ * or visit www.oracle.com if you need additional information or have any
+ * questions.
+ */
+
+import java.awt.Choice;
+import java.awt.EventQueue;
+import java.awt.FlowLayout;
+import java.awt.Frame;
+import java.awt.Point;
+import java.awt.Robot;
+import java.awt.Toolkit;
+import java.awt.event.InputEvent;
+import java.awt.event.KeyEvent;
+import java.awt.event.MouseWheelEvent;
+import java.awt.event.MouseWheelListener;
+
+/*
+ * @test
+ * @bug 6253211
+ * @summary PIT: MouseWheel events not triggered for Choice drop down in XAWT
+ * @requires (os.family == "linux")
+ * @key headful
+ * @run main WheelEventsConsumed
+ */
+
+public class WheelEventsConsumed extends Frame implements MouseWheelListener
+{
+    Robot robot;
+    Choice choice1 = new Choice();
+    Point pt;
+    final static int delay = 100;
+    boolean mouseWheeled = false;
+    final static int OUTSIDE_CHOICE = 1;
+    final static int INSIDE_LIST_OF_CHOICE = 2;
+    final static int INSIDE_CHOICE_COMPONENT = 3;
+    static String toolkit;
+
+    private static volatile WheelEventsConsumed frame = null;
+
+    public static void main(String[] args) throws Exception {
+        toolkit = Toolkit.getDefaultToolkit().getClass().getName();
+        try {
+            EventQueue.invokeAndWait(() -> {
+                frame = new WheelEventsConsumed();
+                frame.initAndShow();
+            });
+            frame.test();
+        } finally {
+            EventQueue.invokeAndWait(() -> {
+                if (frame != null) {
+                    frame.dispose();
+                }
+            });
+        }
+    }
+
+    public void mouseWheelMoved(MouseWheelEvent me) {
+        mouseWheeled = true;
+        System.out.println(me);
+    }
+
+    public void initAndShow() {
+        setTitle("WheelEventsConsumed test");
+        for (int i = 1; i < 10; i++) {
+            choice1.add("item-0" + i);
+        }
+
+        choice1.addMouseWheelListener(this);
+        add(choice1);
+        setLayout(new FlowLayout());
+        setSize(200, 200);
+        setLocationRelativeTo(null);
+        setVisible(true);
+        validate();
+    }
+
+    public void test() {
+        try {
+            robot = new Robot();
+            robot.setAutoWaitForIdle(true);
+            robot.setAutoDelay(50);
+            robot.waitForIdle();
+            robot.delay(delay * 5);
+            testMouseWheel(1, OUTSIDE_CHOICE);
+            robot.delay(delay);
+            testMouseWheel(-1, INSIDE_LIST_OF_CHOICE);
+            robot.delay(delay);
+            testMouseWheel(1, INSIDE_CHOICE_COMPONENT);
+            robot.delay(delay);
+        } catch (Throwable e) {
+            throw new RuntimeException("Test failed. Exception thrown: " + e);
+        }
+    }
+
+    public void testMouseWheel(int amt, int mousePosition) {
+        pt = choice1.getLocationOnScreen();
+        robot.mouseMove(pt.x + choice1.getWidth() / 2, pt.y + choice1.getHeight() / 2);
+
+        robot.mousePress(InputEvent.BUTTON1_DOWN_MASK);
+        robot.delay(50);
+        robot.mouseRelease(InputEvent.BUTTON1_DOWN_MASK);
+        robot.delay(50);
+
+        switch (mousePosition) {
+            case OUTSIDE_CHOICE:
+                robot.mouseMove(pt.x + choice1.getWidth() * 3 / 2, pt.y + choice1.getHeight() / 2);
+                break;
+            case INSIDE_LIST_OF_CHOICE:
+                robot.mouseMove(pt.x + choice1.getWidth() / 2, pt.y + choice1.getHeight() * 4);
+                break;
+            case INSIDE_CHOICE_COMPONENT:
+                robot.mouseMove(pt.x + choice1.getWidth() / 2, pt.y + choice1.getHeight() / 2);
+                break;
+        }
+
+        robot.delay(delay);
+        for (int i = 0; i < 10; i++) {
+            robot.mouseWheel(amt);
+            robot.delay(delay);
+        }
+
+        if (!mouseWheeled) {
+            if (toolkit.equals("sun.awt.windows.WToolkit") && mousePosition == OUTSIDE_CHOICE) {
+                System.out.println("Passed. Separate case on Win32. Choice generated MouseWheel events" + mousePosition);
+            } else {
+                throw new RuntimeException("Test failed. Choice should generate MOUSE_WHEEL events." + mousePosition);
+            }
+        } else {
+            System.out.println("Passed. Choice generated MouseWheel events" + mousePosition);
+        }
+        robot.keyPress(KeyEvent.VK_ESCAPE);
+        robot.delay(10);
+        robot.keyRelease(KeyEvent.VK_ESCAPE);
+        robot.delay(200);
+        mouseWheeled = false;
+    }
+}
diff --git a/test/jdk/java/awt/Focus/ActivateOnProperAppContextTest.java b/test/jdk/java/awt/Focus/ActivateOnProperAppContextTest.java
new file mode 100644
index 0000000000000..ab08398332147
--- /dev/null
+++ b/test/jdk/java/awt/Focus/ActivateOnProperAppContextTest.java
@@ -0,0 +1,245 @@
+/*
+ * Copyright (c) 2006, 2024, Oracle and/or its affiliates. All rights reserved.
+ * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
+ *
+ * This code is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 only, as
+ * published by the Free Software Foundation.
+ *
+ * This code is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+ * version 2 for more details (a copy is included in the LICENSE file that
+ * accompanied this code).
+ *
+ * You should have received a copy of the GNU General Public License version
+ * 2 along with this work; if not, write to the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
+ * or visit www.oracle.com if you need additional information or have any
+ * questions.
+ */
+/*
+* @test
+* @bug 6385277
+* @key headful
+* @summary   Tests that activation happens on correct AppContext.
+* @modules java.desktop/sun.awt
+* @run main ActivateOnProperAppContextTest
+*/
+
+import sun.awt.AppContext;
+import sun.awt.SunToolkit;
+
+import java.awt.Button;
+import java.awt.Component;
+import java.awt.Container;
+import java.awt.Cursor;
+import java.awt.Dimension;
+import java.awt.FlowLayout;
+import java.awt.Frame;
+import java.awt.Label;
+import java.awt.Point;
+import java.awt.Robot;
+import java.awt.Toolkit;
+import java.awt.Window;
+import java.awt.event.InputEvent;
+import java.util.concurrent.atomic.AtomicBoolean;
+
+public class ActivateOnProperAppContextTest {
+    static Robot robot;
+    SunToolkit toolkit;
+
+    ThreadGroup threadGroup = new ThreadGroup("Test_Thread_Group");
+    AppContext appContext;
+    Frame frame;
+    volatile boolean passed = true;
+    AtomicBoolean cond = new AtomicBoolean(false);
+
+    public static void main(String[] args) throws Exception {
+        ActivateOnProperAppContextTest app = new ActivateOnProperAppContextTest();
+        robot = new Robot();
+        app.start();
+    }
+
+    public void start() {
+        toolkit = (SunToolkit)Toolkit.getDefaultToolkit();
+
+        Runnable runnable = new Runnable() {
+                public void run() {
+                    test();
+
+                    synchronized (cond) {
+                        cond.set(true);
+                        cond.notifyAll();
+                    }
+                }
+            };
+
+        Thread thread = new Thread(threadGroup, runnable, "Test Thread");
+
+        synchronized (cond) {
+
+            thread.start();
+
+            while (!cond.get()) {
+                try {
+                    cond.wait();
+                } catch (InterruptedException ie) {
+                    ie.printStackTrace();
+                }
+            }
+        }
+
+        if (passed) {
+            System.out.println("Test passed.");
+        } else {
+            throw new TestFailedException("Test failed!");
+        }
+    }
+
+    void test() {
+        appContext = SunToolkit.createNewAppContext();
+        System.out.println("Created new AppContext: " + appContext);
+
+        frame = new Frame("ActivateOnProperAppContextTest Frame") {
+                public boolean isActive() {
+                    verifyAppContext("Frame.isActive()");
+                    return super.isActive();
+                }
+                public boolean isFocused() {
+                    verifyAppContext("Frame.isFocused()");
+                    return super.isFocused();
+                }
+                public boolean isFocusable() {
+                    verifyAppContext("Frame.isFocusable()");
+                    return super.isFocusable();
+                }
+                public Window getOwner() {
+                    verifyAppContext("Frame.getOwner()");
+                    return super.getOwner();
+                }
+                public boolean isEnabled() {
+                    verifyAppContext("Frame.isEnabled()");
+                    return super.isEnabled();
+                }
+                public boolean isVisible() {
+                    verifyAppContext("Frame.isVisible()");
+                    return super.isVisible();
+                }
+                public Container getParent() {
+                    verifyAppContext("Frame.getParent()");
+                    return super.getParent();
+                }
+                public Cursor getCursor() {
+                    verifyAppContext("Frame.getCursor()");
+                    return super.getCursor();
+                }
+                public Point getLocation() {
+                    verifyAppContext("Frame.getLocation()");
+                    return super.getLocation();
+                }
+                public Point getLocationOnScreen() {
+                    verifyAppContext("Frame.getLocationOnScreen()");
+                    return super.getLocationOnScreen();
+                }
+            };
+        Window window = new Window(frame) {
+                public boolean isFocused() {
+                    verifyAppContext("Window.isFocused()");
+                    return super.isFocused();
+                }
+                public boolean isFocusable() {
+                    verifyAppContext("Window.isFocusable()");
+                    return super.isFocusable();
+                }
+                public Window getOwner() {
+                    verifyAppContext("Window.getOwner()");
+                    return super.getOwner();
+                }
+                public boolean isEnabled() {
+                    verifyAppContext("Window.isEnabled()");
+                    return super.isEnabled();
+                }
+                public boolean isVisible() {
+                    verifyAppContext("Window.isVisible()");
+                    return super.isVisible();
+                }
+                public Container getParent() {
+                    verifyAppContext("Window.getParent()");
+                    return super.getParent();
+                }
+                public Cursor getCursor() {
+                    verifyAppContext("Window.getCursor()");
+                    return super.getCursor();
+                }
+                public Point getLocation() {
+                    verifyAppContext("Window.getLocation()");
+                    return super.getLocation();
+                }
+                public Point getLocationOnScreen() {
+                    verifyAppContext("Window.getLocationOnScreen()");
+                    return super.getLocationOnScreen();
+                }
+            };
+        Button button = new Button("button");
+        Label label = new Label("label");
+
+        window.setLayout(new FlowLayout());
+        window.add(button);
+        window.add(label);
+        window.setLocation(800, 0);
+        window.pack();
+        window.setVisible(true);
+
+        frame.setBounds(800, 100, 100, 50);
+        frame.setVisible(true);
+
+        toolkit.realSync();
+
+        /*
+         * When the label is clicked in the window some of
+         * the owner's public method get called.
+         */
+        clickOn(label);
+    }
+
+    void verifyAppContext(String methodName) {
+        AppContext ac = AppContext.getAppContext();
+        println(methodName + " called on AppContext: " + ac);
+
+        if (ac != appContext) {
+            passed = false;
+            System.err.println("Test failed: " + methodName + " is called on wrong AppContext!");
+            Thread.dumpStack();
+        }
+    }
+
+    void clickOn(Component c) {
+        Point p = c.getLocationOnScreen();
+        Dimension d = c.getSize();
+
+        robot.mouseMove(p.x + (int)(d.getWidth()/2), p.y + (int)(d.getHeight()/2));
+
+        robot.mousePress(InputEvent.BUTTON1_DOWN_MASK);
+        robot.delay(20);
+        robot.mouseRelease(InputEvent.BUTTON1_DOWN_MASK);
+
+        toolkit.realSync();
+    }
+
+    void println(final String msg) {
+        SunToolkit.executeOnEventHandlerThread(frame, new Runnable() {
+                public void run() {
+                    System.out.println(msg);
+                }
+            });
+    }
+}
+
+class TestFailedException extends RuntimeException {
+    TestFailedException(String msg) {
+        super(msg);
+    }
+}
diff --git a/test/jdk/java/awt/Focus/KillFocusTest.java b/test/jdk/java/awt/Focus/KillFocusTest.java
new file mode 100644
index 0000000000000..f61c18971f6f9
--- /dev/null
+++ b/test/jdk/java/awt/Focus/KillFocusTest.java
@@ -0,0 +1,80 @@
+/*
+ * Copyright (c) 2001, 2024, Oracle and/or its affiliates. All rights reserved.
+ * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
+ *
+ * This code is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 only, as
+ * published by the Free Software Foundation.
+ *
+ * This code is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+ * version 2 for more details (a copy is included in the LICENSE file that
+ * accompanied this code).
+ *
+ * You should have received a copy of the GNU General Public License version
+ * 2 along with this work; if not, write to the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
+ * or visit www.oracle.com if you need additional information or have any
+ * questions.
+ */
+/*
+ * @test
+ * @bug 4402942
+ * @summary After deactivation and activation of frame, focus should be restored correctlty
+ * @library /java/awt/regtesthelpers
+ * @build PassFailJFrame
+ * @run main/manual KillFocusTest
+*/
+
+import java.awt.Frame;
+import java.awt.TextField;
+import java.awt.event.FocusEvent;
+import java.awt.event.FocusListener;
+
+public class KillFocusTest {
+
+    private static final String INSTRUCTIONS = """
+         After starting the test you should see \"Test Frame\"
+         with the \"Click me\" text field.
+         Click on this text field and try to type something in it.
+         Make sure that the field receives focus and you can enter text in it.
+         Click on any non-java window.
+         Click on \"Click me\" text field to return focus to it
+         If the caret is in the text field and you are able to type
+         in it then press pass else press fail.""";
+
+    public static void main(String[] args) throws Exception {
+        PassFailJFrame.builder()
+                .title("KillFocusTest Instructions")
+                .instructions(INSTRUCTIONS)
+                .rows((int) INSTRUCTIONS.lines().count() + 2)
+                .columns(35)
+                .testUI(KillFocusTest::createTestUI)
+                .logArea()
+                .build()
+                .awaitAndCheck();
+    }
+
+    private static Frame createTestUI() {
+
+        Frame frame = new Frame("KillFocusTest Frame");
+        TextField textField = new TextField("Click me", 10);
+        textField.addFocusListener(new FocusListener() {
+            public void focusGained(FocusEvent fe) {
+                PassFailJFrame.log("Focus gained");
+            }
+            public void focusLost(FocusEvent fe) {
+                PassFailJFrame.log("Focus lost");
+            }
+        });
+        frame.add(textField);
+        frame.setSize(200, 100);
+        return frame;
+    }
+
+
+}
+
diff --git a/test/jdk/java/awt/Focus/TestDisabledAutoTransfer.java b/test/jdk/java/awt/Focus/TestDisabledAutoTransfer.java
new file mode 100644
index 0000000000000..d7928b7db1c70
--- /dev/null
+++ b/test/jdk/java/awt/Focus/TestDisabledAutoTransfer.java
@@ -0,0 +1,156 @@
+/*
+ * Copyright (c) 2004, 2024, Oracle and/or its affiliates. All rights reserved.
+ * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
+ *
+ * This code is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 only, as
+ * published by the Free Software Foundation.
+ *
+ * This code is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+ * version 2 for more details (a copy is included in the LICENSE file that
+ * accompanied this code).
+ *
+ * You should have received a copy of the GNU General Public License version
+ * 2 along with this work; if not, write to the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
+ * or visit www.oracle.com if you need additional information or have any
+ * questions.
+ */
+/*
+ * @test
+ * @bug 6180261
+ * @summary Test that auto-transfer doesn't happen when there are pending focus requests
+ * @key headful
+ * @run main TestDisabledAutoTransfer
+*/
+
+import java.awt.Button;
+import java.awt.Component;
+import java.awt.Dimension;
+import java.awt.FlowLayout;
+import java.awt.Frame;
+import java.awt.Point;
+import java.awt.Robot;
+import java.awt.event.ActionEvent;
+import java.awt.event.ActionListener;
+import java.awt.event.FocusAdapter;
+import java.awt.event.FocusEvent;
+import java.awt.event.FocusListener;
+import java.awt.event.InputEvent;
+import java.util.concurrent.atomic.AtomicBoolean;
+
+public class TestDisabledAutoTransfer {
+    static Frame frame;
+    static Robot robot;
+    Button b1;
+    Button desired;
+    AtomicBoolean focused = new AtomicBoolean();
+    ActionListener mover;
+    volatile Point loc;
+    volatile Dimension dim;
+
+    public static void main(String[] args) throws Exception {
+        robot = new Robot();
+        try {
+            TestDisabledAutoTransfer test = new TestDisabledAutoTransfer();
+            test.createTestUI();
+            robot.waitForIdle();
+            robot.delay(1000);
+            test.doTest();
+        } finally {
+            if (frame != null) {
+                frame.dispose();
+            }
+        }
+    }
+
+    public void createTestUI() {
+        frame = new Frame("TestDisabledAutoTransfer");
+        frame.setLayout(new FlowLayout());
+        desired = new Button("Desired");
+        FocusAdapter watcher = new FocusAdapter() {
+                public void focusGained(FocusEvent e) {
+                    synchronized(focused) {
+                        focused.set(true);
+                    }
+                }
+            };
+        b1 = new Button("Press to disable");
+        mover = new ActionListener() {
+            public void actionPerformed(ActionEvent e) {
+                desired.requestFocus();
+                    ((Component)e.getSource()).setEnabled(false);
+                }
+            };
+        b1.addFocusListener(watcher);
+        desired.addFocusListener(watcher);
+        frame.add(b1);
+        Button misc = new Button("Next");
+        frame.add(misc);
+        misc.addFocusListener(watcher);
+        frame.add(desired);
+        frame.setSize(200, 200);
+        frame.setLocationRelativeTo(null);
+        frame.setVisible(true);
+        frame.validate();
+
+    }
+
+    public void doTest() {
+
+        loc = b1.getLocationOnScreen();
+        dim = b1.getSize();
+        robot.mouseMove(loc.x + dim.width / 2, loc.y + dim.height / 2);
+        robot.waitForIdle();
+        robot.mousePress(InputEvent.BUTTON1_DOWN_MASK);
+        robot.mouseRelease(InputEvent.BUTTON1_DOWN_MASK);
+        robot.waitForIdle();
+        b1.requestFocus();
+
+        try {
+            synchronized(focused) {
+                if (!focused.get()) {
+                    focused.wait(1000);
+                }
+            }
+        } catch (InterruptedException ie) {
+            throw new RuntimeException("Test was interrupted");
+        }
+
+        if (!focused.get()) {
+            throw new RuntimeException("b1 didn't get focus");
+        }
+        focused.set(false);
+
+        b1.addActionListener(mover);
+        robot.mouseMove(loc.x + dim.width / 2, loc.y + dim.height / 2);
+        robot.waitForIdle();
+        robot.mousePress(InputEvent.BUTTON1_DOWN_MASK);
+        robot.mouseRelease(InputEvent.BUTTON1_DOWN_MASK);
+        robot.waitForIdle();
+
+        try {
+            synchronized(focused) {
+                if (!focused.get()) {
+                    focused.wait(1000);
+                }
+            }
+        } catch (InterruptedException ie) {
+            throw new RuntimeException("Test was interrupted");
+        }
+
+        if (!focused.get()) {
+            throw new RuntimeException("none got focus");
+        }
+
+        if (!desired.isFocusOwner()) {
+            throw new RuntimeException("desired didn't get focus");
+        }
+    }
+
+}
+
diff --git a/test/jdk/java/awt/Focus/TestDisabledAutoTransferSwing.java b/test/jdk/java/awt/Focus/TestDisabledAutoTransferSwing.java
new file mode 100644
index 0000000000000..0793316a989cd
--- /dev/null
+++ b/test/jdk/java/awt/Focus/TestDisabledAutoTransferSwing.java
@@ -0,0 +1,165 @@
+/*
+ * Copyright (c) 2004, 2024, Oracle and/or its affiliates. All rights reserved.
+ * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
+ *
+ * This code is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 only, as
+ * published by the Free Software Foundation.
+ *
+ * This code is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+ * version 2 for more details (a copy is included in the LICENSE file that
+ * accompanied this code).
+ *
+ * You should have received a copy of the GNU General Public License version
+ * 2 along with this work; if not, write to the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
+ * or visit www.oracle.com if you need additional information or have any
+ * questions.
+ */
+/*
+ * @test
+ * @bug 6180261
+ * @summary Test that auto-transfer doesn't happen when there are pending focus requests
+ * @key headful
+ * @run main TestDisabledAutoTransferSwing
+*/
+
+import java.awt.Component;
+import java.awt.Dimension;
+import java.awt.FlowLayout;
+import java.awt.Point;
+import java.awt.Robot;
+import javax.swing.JButton;
+import javax.swing.JFrame;
+import javax.swing.SwingUtilities;
+import java.awt.event.ActionEvent;
+import java.awt.event.ActionListener;
+import java.awt.event.InputEvent;
+import java.awt.event.FocusAdapter;
+import java.awt.event.FocusEvent;
+import java.util.concurrent.atomic.AtomicBoolean;
+
+public class TestDisabledAutoTransferSwing {
+    static JFrame frame;
+    static Robot robot;
+    JButton b1;
+    JButton desired;
+    AtomicBoolean focused = new AtomicBoolean();
+    ActionListener mover;
+    volatile Point loc;
+    volatile Dimension dim;
+
+    public static void main(String[] args) throws Exception {
+        robot = new Robot();
+        try {
+            TestDisabledAutoTransferSwing test = new TestDisabledAutoTransferSwing();
+            SwingUtilities.invokeAndWait(() -> {
+                test.createTestUI();
+            });
+            robot.waitForIdle();
+            robot.delay(1000);
+            test.doTest();
+        } finally {
+            SwingUtilities.invokeAndWait(() -> {
+                if (frame != null) {
+                    frame.dispose();
+                }
+            });
+        }
+    }
+
+    public void createTestUI() {
+        frame = new JFrame("TestDisabledAutoTransferSwing");
+        frame.setLayout (new FlowLayout ());
+        desired = new JButton("Desired");
+        FocusAdapter watcher = new FocusAdapter() {
+                public void focusGained(FocusEvent e) {
+                    synchronized(focused) {
+                        focused.set(true);
+                    }
+                }
+            };
+        b1 = new JButton("Press to disable");
+        mover = new ActionListener() {
+                public void actionPerformed(ActionEvent e) {
+                    desired.requestFocus();
+                    ((Component)e.getSource()).setEnabled(false);
+                }
+            };
+        b1.addFocusListener(watcher);
+        desired.addFocusListener(watcher);
+        frame.add(b1);
+        JButton misc = new JButton("Next");
+        frame.add(misc);
+        misc.addFocusListener(watcher);
+        frame.add(desired);
+        frame.setSize(200, 200);
+        frame.setLocationRelativeTo(null);
+        frame.setVisible(true);
+        frame.validate();
+
+    }
+
+    public void doTest() throws Exception {
+
+        SwingUtilities.invokeAndWait(() -> {
+            loc = b1.getLocationOnScreen();
+            dim = b1.getSize();
+        });
+        robot.mouseMove(loc.x + dim.width / 2, loc.y + dim.height / 2);
+        robot.waitForIdle();
+        robot.mousePress(InputEvent.BUTTON1_DOWN_MASK);
+        robot.mouseRelease(InputEvent.BUTTON1_DOWN_MASK);
+        robot.waitForIdle();
+        SwingUtilities.invokeAndWait(() -> {
+            b1.requestFocus();
+        });
+
+        try {
+            synchronized(focused) {
+                if (!focused.get()) {
+                    focused.wait(2000);
+                }
+            }
+        } catch (InterruptedException ie) {
+            throw new RuntimeException("Test was interrupted");
+        }
+
+        if (!focused.get()) {
+            throw new RuntimeException("b1 didn't get focus");
+        }
+        focused.set(false);
+
+        SwingUtilities.invokeAndWait(() -> {
+            b1.addActionListener(mover);
+        });
+        robot.mouseMove(loc.x + dim.width / 2, loc.y + dim.height / 2);
+        robot.waitForIdle();
+        robot.mousePress(InputEvent.BUTTON1_DOWN_MASK);
+        robot.mouseRelease(InputEvent.BUTTON1_DOWN_MASK);
+        robot.waitForIdle();
+
+        try {
+            synchronized(focused) {
+                if (!focused.get()) {
+                    focused.wait(2000);
+                }
+            }
+        } catch (InterruptedException ie) {
+            throw new RuntimeException("Test was interrupted");
+        }
+
+        if (!focused.get()) {
+            throw new RuntimeException("none got focus");
+        }
+
+        if (!desired.isFocusOwner()) {
+            throw new RuntimeException("desired didn't get focus");
+        }
+    }
+
+}
diff --git a/test/jdk/java/foreign/TestMismatch.java b/test/jdk/java/foreign/TestMismatch.java
index 9549b2508ff64..f50621e34154c 100644
--- a/test/jdk/java/foreign/TestMismatch.java
+++ b/test/jdk/java/foreign/TestMismatch.java
@@ -29,7 +29,9 @@
 
 import java.lang.foreign.Arena;
 import java.util.ArrayList;
+import java.util.Arrays;
 import java.util.List;
+import java.util.Random;
 import java.util.concurrent.atomic.AtomicReference;
 
 import java.lang.foreign.MemorySegment;
@@ -122,6 +124,68 @@ public void testSameValuesStatic(SliceOffsetAndSize ss1, SliceOffsetAndSize ss2)
         }
     }
 
+    @Test
+    public void random() {
+        try (var arena = Arena.ofConfined()) {
+            var rnd = new Random(42);
+            for (int size = 1; size < 64; size++) {
+                // Repeat a fair number of rounds
+                for (int i = 0; i < 147; i++) {
+                    var src = arena.allocate(size);
+                    // The dst segment might be zero to eight bytes longer
+                    var dst = arena.allocate(size + rnd.nextInt(8 + 1));
+                    // Fill the src with random data
+                    for (int j = 0; j < size; j++) {
+                        src.set(ValueLayout.JAVA_BYTE, j, randomByte(rnd));
+                    }
+                    // copy the random data from src to dst
+                    dst.copyFrom(src);
+                    // Fill the rest (if any) of the dst with random data
+                    for (long j = src.byteSize(); j < dst.byteSize(); j++) {
+                        dst.set(ValueLayout.JAVA_BYTE, j, randomByte(rnd));
+                    }
+
+                    if (rnd.nextBoolean()) {
+                        // In this branch, we inject one or more deviating bytes
+                        int beginDiff = rnd.nextInt(size);
+                        int endDiff = rnd.nextInt(beginDiff, size);
+                        for (int d = beginDiff; d <= endDiff; d++) {
+                            byte existing = dst.get(ValueLayout.JAVA_BYTE, d);
+                            // Make sure we never get back the same value
+                            byte mutatedValue;
+                            do {
+                                mutatedValue = randomByte(rnd);
+                            } while (existing == mutatedValue);
+                            dst.set(ValueLayout.JAVA_BYTE, d, mutatedValue);
+                        }
+
+                        // They are not equal and differs in position beginDiff
+                        assertEquals(src.mismatch(dst), beginDiff);
+                        assertEquals(dst.mismatch(src), beginDiff);
+                    } else {
+                        // In this branch, there is no injection
+
+                        if (src.byteSize() == dst.byteSize()) {
+                            // The content matches and they are of equal size
+                            assertEquals(src.mismatch(dst), -1);
+                            assertEquals(dst.mismatch(src), -1);
+                        } else {
+                            // The content matches but they are of different length
+                            // Remember, the size of src is always smaller or equal
+                            // to the size of dst.
+                            assertEquals(src.mismatch(dst), src.byteSize());
+                            assertEquals(dst.mismatch(src), src.byteSize());
+                        }
+                    }
+                }
+            }
+        }
+    }
+
+    static byte randomByte(Random rnd) {
+        return (byte) rnd.nextInt(Byte.MIN_VALUE, Byte.MAX_VALUE + 1);
+    }
+
     @Test(dataProvider = "slices")
     public void testDifferentValues(MemorySegment s1, MemorySegment s2) {
         out.format("testDifferentValues s1:%s, s2:%s\n", s1, s2);
diff --git a/test/jdk/jdk/internal/vm/Continuation/Scoped.java b/test/jdk/jdk/internal/vm/Continuation/Scoped.java
index b034243d596de..1bad8e4aebd00 100644
--- a/test/jdk/jdk/internal/vm/Continuation/Scoped.java
+++ b/test/jdk/jdk/internal/vm/Continuation/Scoped.java
@@ -74,27 +74,27 @@ public void test1() {
 
                         frames = cont.stackWalker().walk(fs -> fs.map(StackWalker.StackFrame::getMethodName).collect(Collectors.toList()));
                         System.out.println("No scope: " + frames);
-                        assertEquals(frames, cont.isDone() ? List.of() : Arrays.asList("yield0", "yield", "lambda$bar$14", "run", "enter0", "enter", "yield0", "run", "bar", "lambda$foo$8", "run", "enter0", "enter", "yield0", "run", "foo", "lambda$test1$0", "run", "enter0", "enter"));
+                        assertEquals(frames, cont.isDone() ? List.of() : Arrays.asList("yield0", "yield", "lambda$bar$0", "run", "enter0", "enter", "yield0", "run", "bar", "lambda$foo$0", "run", "enter0", "enter", "yield0", "run", "foo", "lambda$test1$0", "run", "enter0", "enter"));
 
                         frames = cont.stackWalker(EnumSet.noneOf(StackWalker.Option.class), A).walk(fs -> fs.map(StackWalker.StackFrame::getMethodName).collect(Collectors.toList()));
                         System.out.println("A: " + frames);
-                        assertEquals(frames, cont.isDone() ? List.of() : Arrays.asList("yield0", "yield", "lambda$bar$14", "run", "enter0", "enter", "yield0", "run", "bar", "lambda$foo$8", "run", "enter0", "enter", "yield0", "run", "foo", "lambda$test1$0", "run", "enter0", "enter"));
+                        assertEquals(frames, cont.isDone() ? List.of() : Arrays.asList("yield0", "yield", "lambda$bar$0", "run", "enter0", "enter", "yield0", "run", "bar", "lambda$foo$0", "run", "enter0", "enter", "yield0", "run", "foo", "lambda$test1$0", "run", "enter0", "enter"));
 
                         frames = cont.stackWalker(EnumSet.noneOf(StackWalker.Option.class), B).walk(fs -> fs.map(StackWalker.StackFrame::getMethodName).collect(Collectors.toList()));
                         System.out.println("B: " + frames);
-                        assertEquals(frames, cont.isDone() ? List.of() : Arrays.asList("yield0", "yield", "lambda$bar$14", "run", "enter0", "enter", "yield0", "run", "bar", "lambda$foo$8", "run", "enter0", "enter"));
+                        assertEquals(frames, cont.isDone() ? List.of() : Arrays.asList("yield0", "yield", "lambda$bar$0", "run", "enter0", "enter", "yield0", "run", "bar", "lambda$foo$0", "run", "enter0", "enter"));
 
                         frames = cont.stackWalker(EnumSet.noneOf(StackWalker.Option.class), C).walk(fs -> fs.map(StackWalker.StackFrame::getMethodName).collect(Collectors.toList()));
                         System.out.println("C: " + frames);
-                        assertEquals(frames, cont.isDone() ? List.of() : Arrays.asList("yield0", "yield", "lambda$bar$14", "run", "enter0", "enter"));
+                        assertEquals(frames, cont.isDone() ? List.of() : Arrays.asList("yield0", "yield", "lambda$bar$0", "run", "enter0", "enter"));
 
                         frames = cont.stackWalker(EnumSet.noneOf(StackWalker.Option.class), K).walk(fs -> fs.map(StackWalker.StackFrame::getMethodName).collect(Collectors.toList()));
                         System.out.println("K: " + frames);
-                        assertEquals(frames, cont.isDone() ? List.of() : Arrays.asList("yield0", "yield", "lambda$bar$14", "run", "enter0", "enter", "yield0", "run", "bar", "lambda$foo$8", "run", "enter0", "enter", "yield0", "run", "foo", "lambda$test1$0", "run", "enter0", "enter"));
+                        assertEquals(frames, cont.isDone() ? List.of() : Arrays.asList("yield0", "yield", "lambda$bar$0", "run", "enter0", "enter", "yield0", "run", "bar", "lambda$foo$0", "run", "enter0", "enter", "yield0", "run", "foo", "lambda$test1$0", "run", "enter0", "enter"));
 
                         frames = cont.stackWalker(EnumSet.noneOf(StackWalker.Option.class), null).walk(fs -> fs.map(StackWalker.StackFrame::getMethodName).collect(Collectors.toList()));
                         System.out.println("null: " + frames);
-                        assertEquals(frames, cont.isDone() ? List.of() : Arrays.asList("yield0", "yield", "lambda$bar$14", "run", "enter0", "enter", "yield0", "run", "bar", "lambda$foo$8", "run", "enter0", "enter", "yield0", "run", "foo", "lambda$test1$0", "run", "enter0", "enter"));
+                        assertEquals(frames, cont.isDone() ? List.of() : Arrays.asList("yield0", "yield", "lambda$bar$0", "run", "enter0", "enter", "yield0", "run", "bar", "lambda$foo$0", "run", "enter0", "enter", "yield0", "run", "foo", "lambda$test1$0", "run", "enter0", "enter"));
                 }
                 assertEquals(res.get(), 2);
         }
@@ -119,23 +119,23 @@ static String bar(long b) {
 
                         List<String> frames = StackWalker.getInstance().walk(fs -> fs.map(StackWalker.StackFrame::getMethodName).collect(Collectors.toList()));
 
-                        assertEquals(frames.subList(0, 18), Arrays.asList("lambda$bar$14", "run", "enter0", "enter", "run", "bar", "lambda$foo$8", "run", "enter0", "enter", "run", "foo", "lambda$test1$0", "run", "enter0", "enter", "run", "test1"));
+                        assertEquals(frames.subList(0, 18), Arrays.asList("lambda$bar$0", "run", "enter0", "enter", "run", "bar", "lambda$foo$0", "run", "enter0", "enter", "run", "foo", "lambda$test1$0", "run", "enter0", "enter", "run", "test1"));
 
                         frames = StackWalkerHelper.getInstance(C).walk(fs -> fs.map(StackWalker.StackFrame::getMethodName).collect(Collectors.toList()));
 
-                        assertEquals(frames, Arrays.asList("lambda$bar$14", "run", "enter0", "enter"));
+                        assertEquals(frames, Arrays.asList("lambda$bar$0", "run", "enter0", "enter"));
 
                         frames = StackWalkerHelper.getInstance(B).walk(fs -> fs.map(StackWalker.StackFrame::getMethodName).collect(Collectors.toList()));
 
-                        assertEquals(frames, Arrays.asList("lambda$bar$14", "run", "enter0", "enter", "run", "bar", "lambda$foo$8", "run", "enter0", "enter"));
+                        assertEquals(frames, Arrays.asList("lambda$bar$0", "run", "enter0", "enter", "run", "bar", "lambda$foo$0", "run", "enter0", "enter"));
 
                         frames = StackWalkerHelper.getInstance(A).walk(fs -> fs.map(StackWalker.StackFrame::getMethodName).collect(Collectors.toList()));
 
-                        assertEquals(frames, Arrays.asList("lambda$bar$14", "run", "enter0", "enter", "run", "bar", "lambda$foo$8", "run", "enter0", "enter", "run", "foo", "lambda$test1$0", "run", "enter0", "enter"));
+                        assertEquals(frames, Arrays.asList("lambda$bar$0", "run", "enter0", "enter", "run", "bar", "lambda$foo$0", "run", "enter0", "enter", "run", "foo", "lambda$test1$0", "run", "enter0", "enter"));
 
                         frames = StackWalkerHelper.getInstance(K).walk(fs -> fs.map(StackWalker.StackFrame::getMethodName).collect(Collectors.toList()));
 
-                        assertEquals(frames.subList(0, 18), Arrays.asList("lambda$bar$14", "run", "enter0", "enter", "run", "bar", "lambda$foo$8", "run", "enter0", "enter", "run", "foo", "lambda$test1$0", "run", "enter0", "enter", "run", "test1"));
+                        assertEquals(frames.subList(0, 18), Arrays.asList("lambda$bar$0", "run", "enter0", "enter", "run", "bar", "lambda$foo$0", "run", "enter0", "enter", "run", "foo", "lambda$test1$0", "run", "enter0", "enter", "run", "test1"));
 
                         long r = b+1;
                 });
diff --git a/test/langtools/tools/javac/MethodParameters/LambdaTest.out b/test/langtools/tools/javac/MethodParameters/LambdaTest.out
index b03cc76a97be9..36b91b5e36de1 100644
--- a/test/langtools/tools/javac/MethodParameters/LambdaTest.out
+++ b/test/langtools/tools/javac/MethodParameters/LambdaTest.out
@@ -1,7 +1,7 @@
 class LambdaTest -- 
 LambdaTest.<init>()
 LambdaTest.foo(i)
-LambdaTest.lambda$static$1(arg0)/*synthetic*/
-LambdaTest.lambda$static$0(arg0, arg1)/*synthetic*/
+LambdaTest.lambda$static$0(arg0)/*synthetic*/
+LambdaTest.lambda$static$1(arg0, arg1)/*synthetic*/
 static interface LambdaTest$I -- inner
 LambdaTest$I.m(x)
diff --git a/test/langtools/tools/javac/MethodParameters/LocalClassTest.out b/test/langtools/tools/javac/MethodParameters/LocalClassTest.out
index 3b95739e74a5e..7b0028e1fb0cf 100644
--- a/test/langtools/tools/javac/MethodParameters/LocalClassTest.out
+++ b/test/langtools/tools/javac/MethodParameters/LocalClassTest.out
@@ -1,7 +1,7 @@
 class LocalClassTest$1 -- anon
 LocalClassTest$1.<init>(final this$0/*implicit*/, final j, final val$i/*synthetic*/)
 class LocalClassTest$1CapturingLocal$1 -- anon
-LocalClassTest$1CapturingLocal$1.<init>(final this$0/*implicit*/, final val$val$i/*synthetic*/)
+LocalClassTest$1CapturingLocal$1.<init>(final this$0/*implicit*/, final val$i/*synthetic*/)
 LocalClassTest$1CapturingLocal$1.test()
 class LocalClassTest$1CapturingLocal -- inner
 LocalClassTest$1CapturingLocal.<init>(final this$0/*implicit*/, final j, final val$i/*synthetic*/)
diff --git a/test/langtools/tools/javac/T8019486/WrongLNTForLambdaTest.java b/test/langtools/tools/javac/T8019486/WrongLNTForLambdaTest.java
index 24c9b81c96471..4ddc40fa231a2 100644
--- a/test/langtools/tools/javac/T8019486/WrongLNTForLambdaTest.java
+++ b/test/langtools/tools/javac/T8019486/WrongLNTForLambdaTest.java
@@ -136,15 +136,15 @@ void run() throws Exception {
         checkClassFile(new File(Paths.get(System.getProperty("user.dir"),
                 "Foo.class").toUri()), "lambda$bar$0", simpleLambdaExpectedLNT);
         checkClassFile(new File(Paths.get(System.getProperty("user.dir"),
-                "Foo.class").toUri()), "lambda$variablesInLambdas$1", lambdaWithVarsExpectedLNT);
+                "Foo.class").toUri()), "lambda$variablesInLambdas$0", lambdaWithVarsExpectedLNT);
         checkClassFile(new File(Paths.get(System.getProperty("user.dir"),
                 "Foo$1FooBar.class").toUri()), "run", insideLambdaWithVarsExpectedLNT);
         checkClassFile(new File(Paths.get(System.getProperty("user.dir"),
-                "Foo.class").toUri()), "lambda$variablesInLambdas$2", lambdaVoid2VoidExpectedLNT);
+                "Foo.class").toUri()), "lambda$variablesInLambdas$1", lambdaVoid2VoidExpectedLNT);
         checkClassFile(new File(Paths.get(System.getProperty("user.dir"),
                 "Foo.class").toUri()), "$deserializeLambda$", deserializeExpectedLNT);
         checkClassFile(new File(Paths.get(System.getProperty("user.dir"),
-                "Foo.class").toUri()), "lambda$variablesInLambdas$3", lambdaBridgeExpectedLNT);
+                "Foo.class").toUri()), "lambda$variablesInLambdas$2", lambdaBridgeExpectedLNT);
         checkClassFile(new File(Paths.get(System.getProperty("user.dir"),
                 "Foo.class").toUri()), "assignLambda", assignmentExpectedLNT);
         checkClassFile(new File(Paths.get(System.getProperty("user.dir"),
diff --git a/test/langtools/tools/javac/classfiles/attributes/EnclosingMethod/EnclosingMethodTest.java b/test/langtools/tools/javac/classfiles/attributes/EnclosingMethod/EnclosingMethodTest.java
index b555014bebe67..f15771840f59a 100644
--- a/test/langtools/tools/javac/classfiles/attributes/EnclosingMethod/EnclosingMethodTest.java
+++ b/test/langtools/tools/javac/classfiles/attributes/EnclosingMethod/EnclosingMethodTest.java
@@ -245,7 +245,6 @@ class EnclosingInitialization {
         // anonymous and local classes in lambda
         @ExpectedEnclosingMethod(
                 info = "EnclosingLambda in EnclosingMethodTest",
-                enclosingMethod = "<init>",
                 enclosingClazz = EnclosingMethodTest.class
         )
         class EnclosingLambda {
@@ -325,7 +324,6 @@ class EnclosingInitialization {
             // anonymous and local classes in lambda
             @ExpectedEnclosingMethod(
                     info = "EnclosingLambda in notEnclosing01",
-                    enclosingMethod = "<init>",
                     enclosingClazz = notEnclosing01.class
             )
             class EnclosingLambda {
@@ -382,7 +380,6 @@ public interface notEnclosing02 {
             // anonymous and local classes in lambda
             @ExpectedEnclosingMethod(
                     info = "EnclosingLambda in notEnclosing02",
-                    enclosingMethod = "<clinit>",
                     enclosingClazz = notEnclosing02.class
             )
             class EnclosingLambda {
@@ -460,7 +457,6 @@ class EnclosingInitialization {
             // anonymous and local classes in lambda
             @ExpectedEnclosingMethod(
                     info = "EnclosingLambda in notEnclosing03",
-                    enclosingMethod = "<init>",
                     enclosingClazz = notEnclosing03.class
             )
             class EnclosingLambda {
@@ -517,7 +513,6 @@ static class VariableInitializer {
             // anonymous and local classes in lambda
             @ExpectedEnclosingMethod(
                     info = "EnclosingLambda in notEnclosing04",
-                    enclosingMethod = "<clinit>",
                     enclosingClazz = notEnclosing04.class
             )
             class EnclosingLambda {
diff --git a/test/langtools/tools/javac/lambda/CaptureVarOrder.java b/test/langtools/tools/javac/lambda/CaptureVarOrder.java
new file mode 100644
index 0000000000000..0bfa061cb4f32
--- /dev/null
+++ b/test/langtools/tools/javac/lambda/CaptureVarOrder.java
@@ -0,0 +1,47 @@
+/*
+ * Copyright (c) 2024, Oracle and/or its affiliates. All rights reserved.
+ * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
+ *
+ * This code is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 only, as
+ * published by the Free Software Foundation.
+ *
+ * This code is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+ * version 2 for more details (a copy is included in the LICENSE file that
+ * accompanied this code).
+ *
+ * You should have received a copy of the GNU General Public License version
+ * 2 along with this work; if not, write to the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
+ * or visit www.oracle.com if you need additional information or have any
+ * questions.
+ */
+
+/*
+ * @test
+ * @bug 8336492
+ * @summary Regression in lambda serialization
+ */
+
+public class CaptureVarOrder {
+    static Object m(String s, int i, Object o) {
+        return new Object() {
+            final byte B = 0;
+            void g() { System.out.println(s + i + B + o); }
+        };
+    }
+
+    static Runnable r(String s, int i, Object o) {
+        final byte B = 0;
+        return () -> System.out.println(s + i + B + o);
+    }
+
+    public static void main(String[] args) throws ReflectiveOperationException {
+        CaptureVarOrder.class.getDeclaredMethod("lambda$r$0", String.class, int.class, Object.class);
+        m("", 1, null).getClass().getDeclaredConstructor(String.class, int.class, Object.class);
+    }
+}
diff --git a/test/langtools/tools/javac/lambda/SerializedLambdaInLocalClass.java b/test/langtools/tools/javac/lambda/SerializedLambdaInLocalClass.java
new file mode 100644
index 0000000000000..c015182f44956
--- /dev/null
+++ b/test/langtools/tools/javac/lambda/SerializedLambdaInLocalClass.java
@@ -0,0 +1,95 @@
+/*
+ * Copyright (c) 2024, Oracle and/or its affiliates. All rights reserved.
+ * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
+ *
+ * This code is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 only, as
+ * published by the Free Software Foundation.
+ *
+ * This code is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+ * version 2 for more details (a copy is included in the LICENSE file that
+ * accompanied this code).
+ *
+ * You should have received a copy of the GNU General Public License version
+ * 2 along with this work; if not, write to the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
+ * or visit www.oracle.com if you need additional information or have any
+ * questions.
+ */
+
+/*
+ * @test
+ * @bug 8336492
+ * @summary Regression in lambda serialization
+ */
+
+import java.io.ByteArrayInputStream;
+import java.io.ByteArrayOutputStream;
+import java.io.IOException;
+import java.io.ObjectInputStream;
+import java.io.ObjectOutputStream;
+import java.io.Serializable;
+import java.util.function.*;
+
+public class SerializedLambdaInLocalClass {
+
+    public static void main(String[] args) {
+        SerializedLambdaInLocalClass s = new SerializedLambdaInLocalClass();
+        s.test(s::f_lambda_in_anon);
+        s.test(s::f_lambda_in_local);
+        s.test(s::f_lambda_in_lambda);
+    }
+
+    void test(IntFunction<Supplier<F>> fSupplier) {
+        try {
+            F f = fSupplier.apply(42).get();
+            var baos = new ByteArrayOutputStream();
+            // write
+            try (var oos = new ObjectOutputStream(baos)) {
+                oos.writeObject(f);
+            }
+            byte[] bytes = baos.toByteArray();
+            var bais = new ByteArrayInputStream(bytes);
+            // read
+            try (var ois = new ObjectInputStream(bais)) {
+                F f2 = (F)ois.readObject();
+                if (f2.getValue() != f.getValue()) {
+                    throw new AssertionError(String.format("Found: %d, expected %d", f2.getValue(), f.getValue()));
+                }
+            }
+        } catch (IOException | ClassNotFoundException ex) {
+            throw new AssertionError(ex);
+        }
+    }
+
+    interface F extends Serializable {
+        int getValue();
+    }
+
+    Supplier<F> f_lambda_in_anon(int x) {
+        return new Supplier<F>() {
+            @Override
+            public F get() {
+                return () -> x;
+            }
+        };
+    }
+
+    Supplier<F> f_lambda_in_local(int x) {
+        class FSupplier implements Supplier<F> {
+            @Override
+            public F get() {
+                return () -> x;
+            }
+        }
+        return new FSupplier();
+    }
+
+    Supplier<F> f_lambda_in_lambda(int x) {
+        return () -> () -> x;
+    }
+}
diff --git a/test/micro/org/openjdk/bench/java/lang/foreign/CopyTest.java b/test/micro/org/openjdk/bench/java/lang/foreign/SegmentBulkCopy.java
similarity index 76%
rename from test/micro/org/openjdk/bench/java/lang/foreign/CopyTest.java
rename to test/micro/org/openjdk/bench/java/lang/foreign/SegmentBulkCopy.java
index 8996b1de11742..22ef139aac08a 100644
--- a/test/micro/org/openjdk/bench/java/lang/foreign/CopyTest.java
+++ b/test/micro/org/openjdk/bench/java/lang/foreign/SegmentBulkCopy.java
@@ -41,21 +41,16 @@
 import java.nio.ByteBuffer;
 import java.util.concurrent.TimeUnit;
 
-import static java.lang.foreign.ValueLayout.*;
-
 @BenchmarkMode(Mode.AverageTime)
 @Warmup(iterations = 5, time = 500, timeUnit = TimeUnit.MILLISECONDS)
 @Measurement(iterations = 10, time = 500, timeUnit = TimeUnit.MILLISECONDS)
 @State(Scope.Thread)
 @OutputTimeUnit(TimeUnit.NANOSECONDS)
 @Fork(value = 3)
-public class CopyTest {
+public class SegmentBulkCopy {
 
-    @Param({"0", "1", "2", "3", "4", "5", "6", "7", "8",
-            "9", "10", "11", "12", "13", "14", "15", "16",
-            "17", "18", "19", "20", "21", "22", "23", "24",
-            "25", "26", "27", "28", "29", "30", "31", "32",
-            "33", "36", "40", "44", "48", "52", "56", "60", "63", "64", "128"})
+    @Param({"2", "3", "4", "5", "6", "7", "8", "64", "512",
+            "4096", "32768", "262144", "2097152", "16777216", "134217728"})
     public int ELEM_SIZE;
 
     byte[] srcArray;
@@ -80,28 +75,37 @@ public void setup() {
     }
 
     @Benchmark
-    public void array_copy() {
+    public void arrayCopy() {
         System.arraycopy(srcArray, 0, dstArray, 0, ELEM_SIZE);
     }
 
     @Benchmark
-    public void heap_segment_copy5Arg() {
+    public void bufferCopy() {
+        dstBuffer.put(srcBuffer);
+    }
+
+    @Fork(value = 3, jvmArgsAppend = {"-Djava.lang.foreign.native.threshold.power.copy=31"})
+    @Benchmark
+    public void heapSegmentCopyJava() {
         MemorySegment.copy(heapSrcSegment, 0, heapDstSegment, 0, ELEM_SIZE);
     }
 
+    @Fork(value = 3, jvmArgsAppend = {"-Djava.lang.foreign.native.threshold.power.copy=0"})
     @Benchmark
-    public void native_segment_copy5Arg() {
-        MemorySegment.copy(nativeSrcSegment, 0, nativeDstSegment, 0, ELEM_SIZE);
+    public void heapSegmentCopyUnsafe() {
+        MemorySegment.copy(heapSrcSegment, 0, heapDstSegment, 0, ELEM_SIZE);
     }
 
+    @Fork(value = 3, jvmArgsAppend = {"-Djava.lang.foreign.native.threshold.power.copy=31"})
     @Benchmark
-    public void heap_segment_copy7arg() {
-        MemorySegment.copy(heapSrcSegment, JAVA_BYTE, 0, heapDstSegment, JAVA_BYTE, 0, ELEM_SIZE);
+    public void nativeSegmentCopyJava() {
+        MemorySegment.copy(nativeSrcSegment, 0, nativeDstSegment, 0, ELEM_SIZE);
     }
 
+    @Fork(value = 3, jvmArgsAppend = {"-Djava.lang.foreign.native.threshold.power.copy=0"})
     @Benchmark
-    public void buffer_copy() {
-        dstBuffer.put(srcBuffer);
+    public void nativeSegmentCopyUnsafe() {
+        MemorySegment.copy(nativeSrcSegment, 0, nativeDstSegment, 0, ELEM_SIZE);
     }
 
 }
diff --git a/test/micro/org/openjdk/bench/java/lang/foreign/TestFill.java b/test/micro/org/openjdk/bench/java/lang/foreign/SegmentBulkFill.java
similarity index 69%
rename from test/micro/org/openjdk/bench/java/lang/foreign/TestFill.java
rename to test/micro/org/openjdk/bench/java/lang/foreign/SegmentBulkFill.java
index 78719f03bc377..95ca722896944 100644
--- a/test/micro/org/openjdk/bench/java/lang/foreign/TestFill.java
+++ b/test/micro/org/openjdk/bench/java/lang/foreign/SegmentBulkFill.java
@@ -48,13 +48,10 @@
 @State(Scope.Thread)
 @OutputTimeUnit(TimeUnit.NANOSECONDS)
 @Fork(value = 3)
-public class TestFill {
+public class SegmentBulkFill {
 
-    @Param({"0", "1", "2", "3", "4", "5", "6", "7",
-            "8", "9", "10", "11", "12", "13", "14", "15",
-            "16", "17", "18", "19", "20", "21", "22", "23",
-            "24", "25", "26", "27", "28", "29", "30", "31",
-            "32", "128", "256", "384", "511", "512"})
+    @Param({"2", "3", "4", "5", "6", "7", "8", "64", "512",
+            "4096", "32768", "262144", "2097152", "16777216", "134217728"})
     public int ELEM_SIZE;
 
     byte[] array;
@@ -73,22 +70,43 @@ public void setup() {
     }
 
     @Benchmark
-    public void arrays_fill() {
+    public void arraysFill() {
         Arrays.fill(array, (byte) 0);
     }
 
+    @Fork(value = 3, jvmArgsAppend = {"-Djava.lang.foreign.native.threshold.power.fill=31"})
     @Benchmark
-    public void heap_segment_fill() {
+    public void heapSegmentFillJava() {
         heapSegment.fill((byte) 0);
     }
 
+    @Fork(value = 3, jvmArgsAppend = {"-Djava.lang.foreign.native.threshold.power.fill=0"})
     @Benchmark
-    public void native_segment_fill() {
+    public void heapSegmentFillUnsafe() {
+        heapSegment.fill((byte) 0);
+    }
+
+    @Fork(value = 3, jvmArgsAppend = {"-Djava.lang.foreign.native.threshold.power.fill=31"})
+    @Benchmark
+    public void nativeSegmentFillJava() {
+        nativeSegment.fill((byte) 0);
+    }
+
+    @Fork(value = 3, jvmArgsAppend = {"-Djava.lang.foreign.native.threshold.power.fill=0"})
+    @Benchmark
+    public void nativeSegmentFillUnsafe() {
         nativeSegment.fill((byte) 0);
     }
 
+    @Fork(value = 3, jvmArgsAppend = {"-Djava.lang.foreign.native.threshold.power.fill=31"})
+    @Benchmark
+    public void unalignedSegmentFillJava() {
+        unalignedSegment.fill((byte) 0);
+    }
+
+    @Fork(value = 3, jvmArgsAppend = {"-Djava.lang.foreign.native.threshold.power.fill=0"})
     @Benchmark
-    public void unaligned_segment_fill() {
+    public void unalignedSegmentFillUnsafe() {
         unalignedSegment.fill((byte) 0);
     }
 
diff --git a/test/micro/org/openjdk/bench/java/lang/foreign/SegmentBulkMismatch.java b/test/micro/org/openjdk/bench/java/lang/foreign/SegmentBulkMismatch.java
new file mode 100644
index 0000000000000..5656b2f6b9f74
--- /dev/null
+++ b/test/micro/org/openjdk/bench/java/lang/foreign/SegmentBulkMismatch.java
@@ -0,0 +1,112 @@
+/*
+ *  Copyright (c) 2024, Oracle and/or its affiliates. All rights reserved.
+ *  DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
+ *
+ *  This code is free software; you can redistribute it and/or modify it
+ *  under the terms of the GNU General Public License version 2 only, as
+ *  published by the Free Software Foundation.
+ *
+ *  This code is distributed in the hope that it will be useful, but WITHOUT
+ *  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ *  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+ *  version 2 for more details (a copy is included in the LICENSE file that
+ *  accompanied this code).
+ *
+ *  You should have received a copy of the GNU General Public License version
+ *  2 along with this work; if not, write to the Free Software Foundation,
+ *  Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ *   Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
+ *  or visit www.oracle.com if you need additional information or have any
+ *  questions.
+ *
+ */
+
+package org.openjdk.bench.java.lang.foreign;
+
+import org.openjdk.jmh.annotations.Benchmark;
+import org.openjdk.jmh.annotations.BenchmarkMode;
+import org.openjdk.jmh.annotations.Fork;
+import org.openjdk.jmh.annotations.Measurement;
+import org.openjdk.jmh.annotations.Mode;
+import org.openjdk.jmh.annotations.OutputTimeUnit;
+import org.openjdk.jmh.annotations.Param;
+import org.openjdk.jmh.annotations.Scope;
+import org.openjdk.jmh.annotations.Setup;
+import org.openjdk.jmh.annotations.State;
+import org.openjdk.jmh.annotations.Warmup;
+
+import java.lang.foreign.Arena;
+import java.lang.foreign.MemorySegment;
+import java.util.Arrays;
+import java.util.Random;
+import java.util.concurrent.TimeUnit;
+
+import static java.lang.foreign.ValueLayout.*;
+
+@BenchmarkMode(Mode.AverageTime)
+@Warmup(iterations = 5, time = 500, timeUnit = TimeUnit.MILLISECONDS)
+@Measurement(iterations = 10, time = 500, timeUnit = TimeUnit.MILLISECONDS)
+@State(Scope.Thread)
+@OutputTimeUnit(TimeUnit.NANOSECONDS)
+@Fork(value = 3)
+public class SegmentBulkMismatch {
+
+    @Param({"2", "3", "4", "5", "6", "7", "8", "64", "512",
+            "4096", "32768", "262144", "2097152", "16777216", "134217728"})
+    public int ELEM_SIZE;
+
+    MemorySegment srcNative;
+    MemorySegment dstNative;
+    byte[] srcArray;
+    byte[] dstArray;
+    MemorySegment srcHeap;
+    MemorySegment dstHeap;
+
+    @Setup
+    public void setup() {
+        // Always use the same alignment regardless of size
+        srcNative = Arena.ofAuto().allocate(ELEM_SIZE,16);
+        dstNative = Arena.ofAuto().allocate(ELEM_SIZE, 16);
+        var rnd = new Random(42);
+        for (int i = 0; i < ELEM_SIZE; i++) {
+            srcNative.set(JAVA_BYTE, i, (byte) rnd.nextInt(Byte.MIN_VALUE, Byte.MAX_VALUE));
+        }
+        dstNative.copyFrom(srcNative);
+        srcArray = srcNative.toArray(JAVA_BYTE);
+        dstArray = dstNative.toArray(JAVA_BYTE);
+        srcHeap = MemorySegment.ofArray(srcArray);
+        dstHeap = MemorySegment.ofArray(dstArray);
+    }
+
+    @Fork(value = 3, jvmArgsAppend = {"-Djava.lang.foreign.native.threshold.power.mismatch=31"})
+    @Benchmark
+    public long nativeSegmentJava() {
+        return srcNative.mismatch(dstNative);
+    }
+
+    @Fork(value = 3, jvmArgsAppend = {"-Djava.lang.foreign.native.threshold.power.mismatch=31"})
+    @Benchmark
+    public long heapSegmentJava() {
+        return srcHeap.mismatch(dstHeap);
+    }
+
+    @Fork(value = 3, jvmArgsAppend = {"-Djava.lang.foreign.native.threshold.power.mismatch=0"})
+    @Benchmark
+    public long nativeSegmentUnsafe() {
+        return srcNative.mismatch(dstNative);
+    }
+
+    @Fork(value = 3, jvmArgsAppend = {"-Djava.lang.foreign.native.threshold.power.mismatch=0"})
+    @Benchmark
+    public long heapSegmentUnsafe() {
+        return srcHeap.mismatch(dstHeap);
+    }
+
+    @Benchmark
+    public long array() {
+        return Arrays.mismatch(srcArray, dstArray);
+    }
+
+}
+