GCC 8/9 vs. LLVM Clang 7/8 Compiler Performance On POWER9 With The Raptor Talos II

Earlier this week I delivered the results of our largest-ever GCC vs. LLVM Clang Linux x86_64 compiler comparison with a dozen systems from various generations of Intel and AMD CPUs and using 62 benchmarks tested on GCC 8/9 and Clang 7/8 releases. In this article the compiler performance is being looked at for the IBM POWER9 architecture with the benchmarks done on a Raptor Computing Systems Talos II workstation running Ubuntu Linux.

All of these POWER9 compiler benchmarks were done using the Raptor Talos II with that being our lone POWER9 system at the moment for testing, which was kindly sent over by Raptor Computing Systems. This Talos II configuration has dual 22-core POWER9 processors yielding 44 cores / 176 threads, 64GB of RAM, Samsung 960 EVO NVMe SSD 500GB, and for this testing was running a daily snapshot of Ubuntu 19.04 PPC64LE with the Linux 4.18 kernel.

The compilers benchmarked were GCC 8.2.0, GCC 9.0.1 using this current GCC9 snapshot as of 20190203, LLVM Clang 7.0.1 stable, and the LLVM Clang 8.0 release candidate state as of this past weekend. All of the compilers were built for POWER9 in their release/optimized (non-debug) configuration. During testing, the CFLAGS/CXXFLAGS were set to "-O3 -mtune-native mcpu=native" for benchmarking.

All of these Linux compiler benchmarks were carried out using the Phoronix Test Suite. Following all of the individual benchmark results is also an overview providing a macro look at the GCC vs. Clang compiler performance on POWER9.