GCC Jetson TX1 GCC 4.8 4.9 5.2 Benchmarking Cortex A57 rev 1 testing NVIDIA Jetson Tegra X1 compiler benchmarking. GCC 4.8 through GCC 5.2.1 stable. Benchmarks by Michael Larabel for a future article on Phoronix.com. GCC 4.8.4 - Stock: Processor: Cortex A57 rev 1 @ 1.91GHz (4 Cores), Motherboard: jetson_tx1, Memory: 4096MB, Disk: 16GB 016G32, Graphics: NVIDIA TEGRA OS: Ubuntu 14.04, Kernel: 3.10.67-g3a5c467 (aarch64), Desktop: Unity 7.2.2, Display Server: X Server 1.15.1, Display Driver: NVIDIA 1.0.0, Compiler: GCC 4.8.4 + CUDA 7.0, File-System: ext4, Screen Resolution: 3840x2160 GCC 4.9.3: Processor: Cortex A57 rev 1 @ 1.91GHz (4 Cores), Motherboard: jetson_tx1, Memory: 4096MB, Disk: 16GB 016G32, Graphics: NVIDIA TEGRA OS: Ubuntu 14.04, Kernel: 3.10.67-g3a5c467 (aarch64), Desktop: Unity 7.2.2, Display Server: X Server 1.15.1, Display Driver: NVIDIA 1.0.0, Compiler: GCC 4.9.3 + CUDA 7.0, File-System: ext4, Screen Resolution: 3840x2160 GCC 5.2.1: Processor: Cortex A57 rev 1 @ 1.91GHz (4 Cores), Motherboard: jetson_tx1, Memory: 4096MB, Disk: 16GB 016G32, Graphics: NVIDIA TEGRA OS: Ubuntu 14.04, Kernel: 3.10.67-g3a5c467 (aarch64), Desktop: Unity 7.2.2, Display Server: X Server 1.15.1, Display Driver: NVIDIA 1.0.0, Compiler: GCC 5.2.1 20151031 + CUDA 7.0, File-System: ext4, Screen Resolution: 3840x2160 NAS Parallel Benchmarks 3.3 Test / Class: BT.A Total Mop/s > Higher Is Better GCC 4.8.4 - Stock . 3769.58 |================================================== GCC 4.9.3 ......... 3577.54 |=============================================== GCC 5.2.1 ......... 2781.28 |===================================== NAS Parallel Benchmarks 3.3 Test / Class: EP.B Total Mop/s > Higher Is Better GCC 4.8.4 - Stock . 49.85 |==================================================== GCC 4.9.3 ......... 49.85 |==================================================== GCC 5.2.1 ......... 42.44 |============================================ NAS Parallel Benchmarks 3.3 Test / Class: LU.A Total Mop/s > Higher Is Better GCC 4.8.4 - Stock . 2191.56 |================================================== GCC 4.9.3 ......... 2018.76 |============================================== GCC 5.2.1 ......... 1196.56 |=========================== NAS Parallel Benchmarks 3.3 Test / Class: SP.A Total Mop/s > Higher Is Better GCC 4.8.4 - Stock . 1626.27 |================================================== GCC 4.9.3 ......... 1542.82 |=============================================== GCC 5.2.1 ......... 1215.11 |===================================== Dolfyn 0.527 Computational Fluid Dynamics Seconds < Lower Is Better GCC 4.8.4 - Stock . 108.79 |================================================= GCC 4.9.3 ......... 109.68 |================================================= GCC 5.2.1 ......... 113.06 |=================================================== FFTE 5.0 Test: N=64, 1D Complex FFT Routine MFLOPS > Higher Is Better GCC 4.8.4 - Stock . 2005.31 |================================================= GCC 4.9.3 ......... 2041.50 |================================================== GCC 5.2.1 ......... 2040.03 |================================================== FFTW 3.3.4 Build: Stock - Size: 2D FFT Size 1024 Mflops > Higher Is Better GCC 4.8.4 - Stock . 189.48 |=========================== GCC 4.9.3 ......... 353.92 |=================================================== GCC 5.2.1 ......... 306.19 |============================================ SciMark 2.0 Computational Test: Composite Mflops > Higher Is Better GCC 4.8.4 - Stock . 367.45 |=================================================== GCC 4.9.3 ......... 363.59 |================================================== GCC 5.2.1 ......... 354.41 |================================================= SciMark 2.0 Computational Test: Monte Carlo Mflops > Higher Is Better GCC 4.8.4 - Stock . 187.50 |=================================================== GCC 4.9.3 ......... 186.75 |=================================================== GCC 5.2.1 ......... 188.55 |=================================================== SciMark 2.0 Computational Test: Fast Fourier Transform Mflops > Higher Is Better GCC 4.8.4 - Stock . 37.94 |=============================== GCC 4.9.3 ......... 63.60 |==================================================== GCC 5.2.1 ......... 57.86 |=============================================== SciMark 2.0 Computational Test: Sparse Matrix Multiply Mflops > Higher Is Better GCC 4.8.4 - Stock . 463.17 |=================================================== GCC 4.9.3 ......... 458.87 |=================================================== GCC 5.2.1 ......... 443.31 |================================================= SciMark 2.0 Computational Test: Dense LU Matrix Factorization Mflops > Higher Is Better GCC 4.8.4 - Stock . 700.46 |=================================================== GCC 4.9.3 ......... 680.43 |================================================== GCC 5.2.1 ......... 654.41 |================================================ SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation Mflops > Higher Is Better GCC 4.8.4 - Stock . 448.16 |=================================================== GCC 4.9.3 ......... 428.29 |================================================= GCC 5.2.1 ......... 427.92 |================================================= John The Ripper 1.8.0 Test: MD5 Real C/S > Higher Is Better GCC 4.8.4 - Stock . 32229 |==================================================== GCC 4.9.3 ......... 30269 |================================================= GCC 5.2.1 ......... 21894 |=================================== VP8 libvpx Encoding 1.3.0 vpxenc Frames Per Second > Higher Is Better GCC 4.8.4 - Stock . 9.20 |===================================================== GCC 4.9.3 ......... 7.82 |============================================= GCC 5.2.1 ......... 7.27 |========================================== 7-Zip Compression 9.20.1 Compress Speed Test MIPS > Higher Is Better GCC 4.8.4 - Stock . 4294 |===================================================== GCC 4.9.3 ......... 4171 |=================================================== GCC 5.2.1 ......... 3960 |================================================= C-Ray 1.1 Total Time Seconds < Lower Is Better GCC 4.8.4 - Stock . 93.58 |=============================================== GCC 4.9.3 ......... 101.35 |=================================================== GCC 5.2.1 ......... 96.67 |================================================= Parallel BZIP2 Compression 1.1.6 256MB File Compression Seconds < Lower Is Better GCC 4.8.4 - Stock . 35.38 |================================================ GCC 4.9.3 ......... 32.96 |============================================= GCC 5.2.1 ......... 38.09 |==================================================== Smallpt 1.0 Global Illumination Renderer; 100 Samples Seconds < Lower Is Better GCC 4.8.4 - Stock . 614 |============================================== GCC 4.9.3 ......... 643 |================================================ GCC 5.2.1 ......... 718 |====================================================== N-Queens 1.0 Elapsed Time Seconds < Lower Is Better GCC 4.8.4 - Stock . 119.50 |============================================== GCC 4.9.3 ......... 122.10 |=============================================== GCC 5.2.1 ......... 132.34 |===================================================