NASM changes JNZ to JNE while assembling? Why?

Question

I have a piece of code which uses JNZ. When I assemble and link the binary, I see my JNZ is replaces with a JNE. I understand that both of them fundamentally are the same. But then why does NASM change it?

Also, is there any config option available to stop this change from happening while assembling?

Take some time to read CPU documentation. JE and JZ are the same thing. And so are JNE and JNZ. — Alexey Frunze, Feb 11 '13 at 06:57
Why do we have synonyms, aliases, nicknames and such? They serve some purpose. What is more natural to think of, `Equality` or `Zero` after `ADD AX, 1`? Now, answer the same question for `CMP AX, 1`. — Alexey Frunze, Feb 11 '13 at 09:33
I'm simply amazed to see lack of effort where you could have easily found the answer yourself. — Alexey Frunze, Feb 11 '13 at 12:22
You come on. :) You'll be a better programmer if you are proactive. — Alexey Frunze, Feb 11 '13 at 12:38
because [they're the same](http://stackoverflow.com/questions/14267081/difference-between-je-jne-and-jz-jnz) — phuclv, Apr 22 '15 at 02:01

Andreas Fester · Accepted Answer · 2013-02-11T06:43:12.840

11

I understand that both of them fundamentally are the same

JNE and JNZ have the same opcodes (0x75 for short jumps and 0x0f 0x85 for near jumps), so the assembler will create the same machine code for both of them.

When disassembling, the disassembler does not known anymore which one was used in the source and it has to take one of them.

Also, is there any config option available to stop this change from happening while assembling?

No, because it is not a real "replacement" - JNE and JNZ are simply different mnemonics for the same opcodes.

edited Feb 11 '13 at 06:43

answered Feb 11 '13 at 06:35

Andreas Fester

36,091
7
95
123

My next question is then why need them both? – ST-User Feb 11 '13 at 09:22
4

You can use them to make your code more readable and easier to understand - use `JNE` followed by a `CMP` instruction to make clear that you are checking for **equality**, and use `JNZ` if you simply want to check the zero flag (e.g. after a `SUB` instruction to check if the result is zero or not) – Andreas Fester Feb 11 '13 at 09:28

score 2 · Answer 2 · answered Feb 11 '13 at 06:38

JNZ and JNE have exactly the same encoding (refer to Intel® 64 and IA-32 Architectures Software Developer’s Manual Vol. 2A 3-419). So whichever you use in the assembler, the disassembler would pick one and use the same notation throughout in the disassembled code.

NASM changes JNZ to JNE while assembling? Why?

2 Answers2