Prerequisites - Prerequisites

Install the following software packages before installing AOCC.

Table 1. AOCC Prerequisites
Package Name	Version(s)	Notes
libstdc++	6 or later	GNUStandard C++ Library Note: libstdc++9 or later is required for C++20 standard.
libncurses-dev	5.9 or later	Provides libtinfo, which is a low level terminfo library
zlib	1.2.7 or later	Compression library
Libxml2	2 or later	Parses the XML documents
libquadmath	4.8 or later	GCC Quad-Precision Math Library
python	3.x	Python library

Note:

For better performance, it is recommended to use the latest versions of glibc and Binutils.
AOCC compiler binaries are suitable only for the Linux® systems with glibc version 2.28 or later.
On RHEL, if there are multiple versions of gcc-toolset-N installed in the system, ensure that the libquadmath.so file exists in the respective directory where GCC toolset is installed. For further details, refer to the RHEL documentation.

Important: Troubleshooting Tip: Starting with LLVM version 16, libc++ defaults to the C++17 standard. If you are compiling C++ code written for earlier standards using libc++ from LLVM 16 or newer, you may encounter compatibility issues such as missing members or unresolved symbols. To avoid these errors, build your code with the flag -std=c++14.

Prerequisites for AORs

Table 2. AOR Prerequisites
Package Name	Version(s)	Notes
Python	3.8 or later	Python library
Python3-pip	23.1.2 or later	Package management utility
pip3 package Pygments	2.15.1 or later	Generic syntax highlighter utility
pip3 package PyYAML	6.0 or later	YAML package for Python
pip3 package Psutil	7.0.0 or later	Cross platform library

Note: Install the pip packages compatible with the Python version used with AOR. Here are a few examples.

Python 3.8 is compatible with Python3-pip 23.1.2, pip3 package Pygments 2.15.1, and pip3 package PyYAML 6.0
Python 3.12 is compatible with Python3-pip 25.0.1, pip3 package Pygments 2.19.1, and pip3 package PyYAML 6.0.1

Prerequisites - Prerequisites - 57222

AOCC User Guide