[compiler] Make SFI "optimize" flag a "tier up" flag

src/builtins/arm64/builtins-arm64.cc

Index: src/builtins/arm64/builtins-arm64.cc
diff --git a/src/builtins/arm64/builtins-arm64.cc b/src/builtins/arm64/builtins-arm64.cc
index 2d7c55f651aa1f8072d748ea502d8d2eca6c8b23..cd0f18fe8d1dc2947042a8951704ef27ca905b04 100644
--- a/src/builtins/arm64/builtins-arm64.cc
+++ b/src/builtins/arm64/builtins-arm64.cc
@@ -1475,11 +1475,10 @@ void Builtins::Generate_CompileLazy(MacroAssembler* masm) {
   __ Ldr(entry,
          FieldMemOperand(closure, JSFunction::kSharedFunctionInfoOffset));
   // Is the shared function marked for optimization?
-  __ Ldrb(temp,
-          FieldMemOperand(
-              entry, SharedFunctionInfo::kWasMarkedForOptimizationByteOffset));
+  __ Ldrb(temp, FieldMemOperand(
+                    entry, SharedFunctionInfo::kWasMarkedForTierUpByteOffset));
   __ TestAndBranchIfAnySet(
-      temp, 1 << SharedFunctionInfo::kWasMarkedForOptimizationBitWithinByte,
+      temp, 1 << SharedFunctionInfo::kWasMarkedForTierUpBitWithinByte,
       &gotta_call_runtime);
   // Is the full code valid?
   __ Ldr(entry, FieldMemOperand(entry, SharedFunctionInfo::kCodeOffset));