GCC vs. LLVM Clang On NVIDIA's Tegra K1 Quad-Core Cortex-A15
Recently I posted new benchmarks showing LLVM's Clang compiler performing well against GCC from AMD's x86-based Athlon APUs with the performance of the resulting binaries being quite fast but not without some blemishes for both of these open-source compilers. In seeing how the compiler race is doing in the ARM space with many ARM vendors taking interest in LLVM/Clang, here's some fresh benchmarks of both compilers on NVIDIA's Tegra K1 SoC found by the Jetson TK1 development board.
Within this article are benchmarks of GCC and LLVM/Clang as packaged for Ubuntu 14.04 LTS ARMv7, which means GCC 4.8.2 and LLVM Clang 3.4. NVIDIA's spin of Ubuntu 14.04 LTS for the Jetson TK1 ARM board uses the Linux 3.10 kernel for the Tegra K1 with its quad-core Cortex-A15 processor plus Kepler graphics.
During benchmarking, the compiler flags and other settings were maintained the same during testing with the only change between setting up GCC and Clang from the Ubuntu 14.04 Trusty Tahr archive was changing the compiler in use and then rebuilding all of the used test profiles within the Phoronix Test Suite.
While in this article we are just looking at the GCC 4.8 and LLVM/Clang 3.4 compiler performance given that's what most will be using since it's the versions officially supplied by Ubuntu 14.04 LTS, in another upcoming article I will also take a look at the ARM Cortex-A15 / NVIDIA Tegra K1 performance when using the latest GCC 4.9.0 release and the Clang 3.5 development code.