CUDA TK1
ARMv7 rev 3 testing with a jetson-tk1 and GK20A/NullRM/AXI on Ubuntu 14.04 via the Phoronix Test Suite.


Nbody TK1: 

	Processor: ARMv7 rev 3 @ 2.32GHz (4 Cores), Motherboard: jetson-tk1, Chipset: NVIDIA TegraK1, Memory: 2048MB, Disk: 16GB SEM16G, Graphics: GK20A/NullRM/AXI, Monitor: DELL E1909W, Network: Realtek RTL8111/8168/8411

	OS: Ubuntu 14.04, Kernel: 3.10.40-g8c4516e (armv7l), Desktop: Unity 7.2.6, Display Server: X Server 1.15.1, Display Driver: NVIDIA 21.1, OpenGL: 4.3.0, Compiler: GCC 4.8.4 + CUDA 6.0, File-System: ext4, Screen Resolution: 1440x900


CUDA Mini-Nbody 2015-11-10
Test: Original
Seconds < Lower Is Better
Nbody TK1 . 8.28 |=============================================================


CUDA Mini-Nbody 2015-11-10
Test: Cache Blocking
Seconds < Lower Is Better
Nbody TK1 . 8.57 |=============================================================


CUDA Mini-Nbody 2015-11-10
Test: Loop Unrolling
Seconds < Lower Is Better
Nbody TK1 . 10.54 |============================================================


CUDA Mini-Nbody 2015-11-10
Test: SOA Data Layout
Seconds < Lower Is Better
Nbody TK1 . 8.61 |=============================================================


CUDA Mini-Nbody 2015-11-10
Test: Flush Denormals To Zero
Seconds < Lower Is Better
Nbody TK1 . 8.58 |=============================================================