Prepare your AArch64 environment and verify LLVM tool availability

Set up your environment

On your AArch64 Linux machine, create a working directory and enter it:

    

        
        
mkdir pgo-example
cd pgo-example

    

Download the bsort.cpp source file from the Arm Learning Paths repository:

    

        
        
wget https://raw.githubusercontent.com/ArmDeveloperEcosystem/arm-learning-paths/main/content/learning-paths/servers-and-cloud-computing/bolt/bsort.cpp

    

Create directories for the generated files:

    

        
        
mkdir -p out prof

    

The out directory stores object files and binaries. The prof directory stores raw and converted profile data.

If LLVM isn’t already installed, follow the LLVM toolchain for Linux on Arm install guide before continuing.

Verify LLVM tool availability

Confirm that the machine uses the AArch64 architecture:

    

        
        
uname -m

    

The expected output is:

    

        
        aarch64

        
    

Check that the required LLVM tools are available:

    

        
        
clang++ --version
ld.lld --version
llvm-bcanalyzer --version
llvm-profdata --version
llvm-profgen --version
llvm-readelf --version

    

For sample-based profile-guided optimization (S-PGO), also check that perf is available:

    

        
        
perf --version
uname -r

    

The S-PGO workflow uses the Arm Branch Record Buffer Extension (BRBE) and needs Linux kernel 6.17 or later.

The version commands confirm that the programs are in your PATH. They don’t confirm that the processor implements BRBE or that you have permission to access performance events. The profile collection step checks those requirements when it runs perf record.

What you’ve accomplished and what’s next

You’ve created the working directory, downloaded the example source, and verified that the required tools are available.

Next, you’ll build the example with Thin-LTO and Full-LTO.