Question 380266 in Xcode Question posted in
Whether you're new to Xcode or an experienced developer, our archive has everything you need to know about this integrated development environment (IDE). From basic functionalities to advanced features, our archive covers a wide range of Xcode-related questions and answers. Browse our archive now and find solutions to your Xcode questions, and take your app development skills to the next level

Xcode Apple Silicon not comping ARM64 half precision neon instructions: Invalid operand for instruction

To date I have had no issue compiling and running complex ARM Neon assembly language routines in Xcode/CLANG, and the Apple M1 supposedly supports ARMv8.4.

But – when I try to use half precision with the following syntax:
" fadd v2.8h, v0.8h, v1.8h nt"
I get the compiler error "Invalid Operand for instruction".
Adding -mfpu=neon-fp16 to the compiler options didn’t help. Hoping it just uses a different syntax?

If anyone has achieved .8h instructions compiling in Xcode, I would be very grateful for any tips. Thank you.

Tags:

2

Answers


  1. Chosen as BEST ANSWER
    0

    I think I've solved the issue - or at least half of the issue - I can now get it to compile. There were two key points:

    1. The compiler flag to add is JUST -march=armv8.2-a+fp16 and nothing else. I was trying too many things.

    2. In XCode the way to add the flags is per C file. Specifically... Click on your project go to the Build Phases Tab. Click on your target and you should see a list of the source files. Click on each relevant source file and add -march=armv8.2-a+fp16

    After I did this, it compiled. I will add another comment if I find more info.


  2. 0

    That’s wonderful news. :). Maybe over time this is becoming standard.

    Login or Signup to reply.
Please signup or login to give your own answer.

Back To Top
Search