[interpreter] Optimize return bytecodes on arm

Tweaks AdvanceBytecodeOffsetOrReturn so that the sequence of (cmp,beq)+
instructions is converted to (cmp, cmpne+, beq) saving an instruction
for every return bytecode. In reality this just saves a single
instruction.

Bug: v8:9771
Change-Id: I7cf2d5ae27ff5495808792aa4c953b97c2bb5b71
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1853246
Commit-Queue: Dan Elphick <delphick@chromium.org>
Reviewed-by: Santiago Aboy Solanes <solanes@chromium.org>
Cr-Commit-Position: refs/heads/master@{#64232}

src/builtins/arm/builtins-arm.cc

@@ -992,12 +992,18 @@ static void AdvanceBytecodeOffsetOrReturn(MacroAssembler* masm,
   __ bind(&process_bytecode);
 
 // Bailout to the return label if this is a return bytecode.
-#define JUMP_IF_EQUAL(NAME)                                                    \
-  __ cmp(bytecode, Operand(static_cast<int>(interpreter::Bytecode::k##NAME))); \
-  __ b(if_return, eq);
+
+  // Create cmp, cmpne, ..., cmpne to check for a return bytecode.
+  Condition flag = al;
+#define JUMP_IF_EQUAL(NAME)                                                   \
+  __ cmp(bytecode, Operand(static_cast<int>(interpreter::Bytecode::k##NAME)), \
+         flag);                                                               \
+  flag = ne;
   RETURN_BYTECODE_LIST(JUMP_IF_EQUAL)
 #undef JUMP_IF_EQUAL
 
+  __ b(if_return, eq);
+
   // Otherwise, load the size of the current bytecode and advance the offset.
   __ ldr(scratch1, MemOperand(bytecode_size_table, bytecode, LSL, 2));
   __ add(bytecode_offset, bytecode_offset, scratch1);