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

src/builtins/mips64/builtins-mips64.cc

Index: src/builtins/mips64/builtins-mips64.cc
diff --git a/src/builtins/mips64/builtins-mips64.cc b/src/builtins/mips64/builtins-mips64.cc
index ea94066563f958fa9727d578ed1c91aa3540f773..a8f1cccc0f7c0bc952b6c40d151603c7e3a0171c 100644
--- a/src/builtins/mips64/builtins-mips64.cc
+++ b/src/builtins/mips64/builtins-mips64.cc
@@ -1472,13 +1472,11 @@ void Builtins::Generate_CompileLazy(MacroAssembler* masm) {
   __ pop(new_target);
   __ pop(argument_count);
   __ ld(entry, FieldMemOperand(closure, JSFunction::kSharedFunctionInfoOffset));
-  // Is the shared function marked for optimization?
-  __ lbu(a5,
-         FieldMemOperand(
-             entry, SharedFunctionInfo::kWasMarkedForOptimizationByteOffset));
-  __ And(
-      a5, a5,
-      Operand(1 << SharedFunctionInfo::kWasMarkedForOptimizationBitWithinByte));
+  // Is the shared function marked for tier up?
+  __ lbu(a5, FieldMemOperand(entry,
+                             SharedFunctionInfo::kMarkedForTierUpByteOffset));
+  __ And(a5, a5,
+         Operand(1 << SharedFunctionInfo::kMarkedForTierUpBitWithinByte));
   __ Branch(&gotta_call_runtime_no_stack, ne, a5, Operand(zero_reg));
   // Is the full code valid?
   __ ld(entry, FieldMemOperand(entry, SharedFunctionInfo::kCodeOffset));