ARM64 gfortran performance comparison
Fortran compiler tuning on aarch64


5.4.0: 

	Processor: Unknown @ 1.50GHz (4 Cores), Memory: 2048MB, Disk: 16GB NCard + 32GB 00000

	OS: Ubuntu 16.04, Kernel: 3.14.65-61 (aarch64), Compiler: GCC 5.4.0 20160609 + LLVM 3.8.0, File-System: ext4, Screen Resolution: 1280x1440

6.2.1: 

	Processor: Unknown @ 1.50GHz (4 Cores), Memory: 2048MB, Disk: 16GB NCard + 32GB 00000

	OS: Ubuntu 16.04, Kernel: 3.14.65-61 (aarch64), Compiler: GCC 6.2.1 20160901 + LLVM 3.8.0, File-System: ext4, Screen Resolution: 1280x1440

7.0.0: 

	Processor: Unknown @ 1.50GHz (4 Cores), Memory: 2048MB, Disk: 16GB NCard + 32GB 00000

	OS: Ubuntu 16.04, Kernel: 3.14.65-61 (aarch64), Compiler: GCC 7.0.0 20160923 + LLVM 3.8.0, File-System: ext4, Screen Resolution: 1280x1440

7.0.0/A53: 

	Processor: Unknown @ 1.50GHz (4 Cores), Memory: 2048MB, Disk: 16GB NCard + 32GB 00000

	OS: Ubuntu 16.04, Kernel: 3.14.65-61 (aarch64), Compiler: GCC 7.0.0 20160923 + LLVM 3.8.0, File-System: ext4, Screen Resolution: 1280x1440

6.2.1/A53: 

	Processor: Unknown @ 1.50GHz (4 Cores), Memory: 2048MB, Disk: 16GB NCard + 32GB 00000

	OS: Ubuntu 16.04, Kernel: 3.14.65-61 (aarch64), Compiler: GCC 6.2.1 20160901 + LLVM 3.8.0, File-System: ext4, Screen Resolution: 1280x1440

7.0.1: 

	Processor: Unknown @ 1.54GHz (4 Cores), Motherboard: Amlogic, Memory: 2048MB, Disk: 16GB NCard + 32GB 00000

	OS: Ubuntu 16.04, Kernel: 3.14.79-vegas95 (aarch64), Compiler: GCC 7.0.1 20170306 + Clang 3.8.0-2ubuntu4 + LLVM 3.8.0, File-System: ext4, Screen Resolution: 1280x1440

7.0.1/A57 tuned: 

	Processor: Unknown @ 1.54GHz (4 Cores), Motherboard: Amlogic, Memory: 2048MB, Disk: 16GB NCard + 32GB 00000

	OS: Ubuntu 16.04, Kernel: 3.14.79-vegas95 (aarch64), Compiler: GCC 7.0.1 20170307 + Clang 3.8.0-2ubuntu4 + LLVM 3.8.0, File-System: ext4, Screen Resolution: 1280x1440


NAS Parallel Benchmarks 3.3
Test / Class: LU.A
Total Mop/s > Higher Is Better
5.4.0 ........... 753.88 |==============================================
6.2.1 ........... 748.24 |==============================================
7.0.0 ........... 754.35 |==============================================
7.0.0/A53 ....... 861.02 |=====================================================
6.2.1/A53 ....... 857.04 |=====================================================
7.0.1 ........... 754.51 |==============================================
7.0.1/A57 tuned . 858.70 |=====================================================


Dolfyn 0.527
Computational Fluid Dynamics
Seconds < Lower Is Better
5.4.0 ........... 272.53 |==========================================
6.2.1 ........... 345.05 |=====================================================
7.0.0 ........... 268.12 |=========================================
7.0.0/A53 ....... 268.24 |=========================================
6.2.1/A53 ....... 269.17 |=========================================
7.0.1 ........... 268.37 |=========================================
7.0.1/A57 tuned . 250.38 |======================================


FFTE 5.0
Test: N=64, 1D Complex FFT Routine
MFLOPS > Higher Is Better
5.4.0 ........... 1505.70 |====================================================
6.2.1 ........... 1510.71 |====================================================
7.0.0 ........... 1443.32 |=================================================
7.0.0/A53 ....... 1442.95 |=================================================
6.2.1/A53 ....... 1441.56 |=================================================
7.0.1 ........... 1517.95 |====================================================
7.0.1/A57 tuned . 1501.56 |===================================================