OMAP4660 ARM LLVM-Clang On Ubuntu 12.04
Phoronix benchmarks of GCC 4.6 and LLVM/Clang 3.0 from Precise repositories on Ubuntu 12.04 LTS snapshot. Both compilers were using same test profile defaults for compile flags. Testing for a future article on Phoronix.com by Michael Larabel.


GCC 4.6: 

	Processor: ARMv7 rev 10 (2 Cores), Motherboard: OMAP4 Panda board, Memory: 912MB, Disk: 16GB SD16G, Graphics: Texas Instruments OMAP

	OS: Ubuntu 12.04, Kernel: 3.2.0-1404-omap4 (armv7l), Desktop: Unity 2D, Display Server: X Server 1.10.4, Display Driver: fbdev 0.4.2, OpenGL: 2.1 Mesa 7.11, Compiler: GCC 4.6, File-System: ext4, Screen Resolution: 1024x768

LLVM-Clang 3.0: 

	Processor: ARMv7 rev 10 (2 Cores), Motherboard: OMAP4 Panda board, Memory: 912MB, Disk: 16GB SD16G, Graphics: Texas Instruments OMAP

	OS: Ubuntu 12.04, Kernel: 3.2.0-1404-omap4 (armv7l), Desktop: Unity 2D, Display Server: X Server 1.10.4, Display Driver: fbdev 0.4.2, OpenGL: 2.1 Mesa 7.11, Compiler: Clang 4.2.1 + GCC 4.6, File-System: ext4, Screen Resolution: 1024x768


CacheBench
Test: Read
MB/s > Higher Is Better
GCC 4.6 ........ 1015.30 |=====================================================
LLVM-Clang 3.0 . 162.32  |========


CacheBench
Test: Write
MB/s > Higher Is Better
GCC 4.6 ........ 1443.61 |===========================================
LLVM-Clang 3.0 . 1783.72 |=====================================================


CacheBench
Test: Read / Modify / Write
MB/s > Higher Is Better
GCC 4.6 ........ 1480.92 |=====================================================
LLVM-Clang 3.0 . 243.70  |=========


C-Ray 1.1
Total Time
Seconds < Lower Is Better
GCC 4.6 ........ 1961.29 |=============
LLVM-Clang 3.0 . 8180.15 |=====================================================


7-Zip Compression 9.20.1
Compress Speed Test
MIPS > Higher Is Better
GCC 4.6 ........ 542 |========================================================
LLVM-Clang 3.0 . 548 |=========================================================


LZMA Compression
256MB File Compression
Seconds < Lower Is Better
GCC 4.6 ........ 1070.19 |================================================
LLVM-Clang 3.0 . 1182.29 |=====================================================


Timed MAFFT Alignment 6.864
Multiple Sequence Alignment
Seconds < Lower Is Better
GCC 4.6 ........ 248.76  |===========
LLVM-Clang 3.0 . 1175.70 |=====================================================


N-Queens 1.0
Elapsed Time
Seconds < Lower Is Better
GCC 4.6 ........ 1420.96 |=====================================================
LLVM-Clang 3.0 . 1222.51 |==============================================


SciMark 2.0
Computational Test: Composite
Mflops > Higher Is Better
GCC 4.6 ........ 65.90 |=======================================================
LLVM-Clang 3.0 . 9.11  |========


SciMark 2.0
Computational Test: Fast Fourier Transform
Mflops > Higher Is Better
GCC 4.6 ........ 15.15 |=======================================================
LLVM-Clang 3.0 . 3.40  |============


SciMark 2.0
Computational Test: Jacobi Successive Over-Relaxation
Mflops > Higher Is Better
GCC 4.6 ........ 121.86 |======================================================
LLVM-Clang 3.0 . 14.47  |======


SciMark 2.0
Computational Test: Monte Carlo
Mflops > Higher Is Better
GCC 4.6 ........ 45.05 |=======================================================
LLVM-Clang 3.0 . 6.52  |========


SciMark 2.0
Computational Test: Sparse Matrix Multiply
Mflops > Higher Is Better
GCC 4.6 ........ 70.84 |=======================================================
LLVM-Clang 3.0 . 11.43 |=========


SciMark 2.0
Computational Test: Dense LU Matrix Factorization
Mflops > Higher Is Better
GCC 4.6 ........ 76.61 |=======================================================
LLVM-Clang 3.0 . 9.74  |=======


TSCP 1.81
AI Chess Performance
Nodes Per Second > Higher Is Better
GCC 4.6 ........ 42620 |=======================================================
LLVM-Clang 3.0 . 37733 |=================================================