Mamona, an OpenEmbedded based platform that I’ve been worked on for the last year, has a new release. The Mamona team are proud to announce the Mamona 0.2 that represents a huge step forward. The main news of this release are:

• Glibc – 2.5
• GCC – 4.1.2
• Full Python 2.5 support
• Enlightenment Window Manager
• USB Networking
• Bluetooth support
• Noemu Packages
• Mamona SDK
• Patched QEMU (user mode emulation) for Mamona SDK
• ARMv5te (N770) and ARMv6 (N8X0 and Omap 3430sdp)
• Machines Supported:
  • Nokia 770
  • Nokia 800
  • Nokia 810
  • Omap 3430sdp
• Web browsers: Firefox 3 and Midori (recommended)
• Virtual Keyboard
• Xephyr support on SDK
• SystemTap enabled kernels for Nokia devices
• Full Emulation
• Network Manager working with dbus API
• Command line interface for Network Manager
• Easier installer, including deb, rpm and tgz “packages”
• Completely rewritten wiki documentation

Check the release notes at http://dev.openbossa.org/trac/mamona/wiki/Releases/0.2/ReleaseNotes.

This week I finished some performance tests of LLVM for ARM. The results were very good. I tested the compilation of two programs: FFMPEG and GTK. Each program was compiled in three ways: gcc, llvm-gcc and llvm-gcc with modifications in the programs build systems to use link-time optimizations. All tests were ran on an Texas Instruments OMAP 1710 board with 32MB of RAM running Debian Linux with kernel 2.6.18-omap1.

In these tests, only the main program was compiled by the llvm-gcc. Any dependency was picked in binary form from the Debian distribution. If we compile all dependencies with llvm-gcc, probably, we can get better results.

Program versions:
gcc – 4.1.2
llvm-gcc – svn revison 40660
ffmpeg – svn revision 9470
gtk – 2.10.12

FFMPEG test
Average time to convert 1000 frames from mpeg4 to mpeg1.
gcc – 336.87s
llvm-gcc – 323.9 s (4% better than gcc)
llvm-gcc with LTO – 323.7s (4% better than gcc)

GTK test
Average time to run perftest (a gtk internal test) with 1000 iterations.
gcc – 198.29s
llvm-gcc – 192.7s (3% better than gcc)
llvm-gcc with LTO – 194.65s (2% better than gcc)

Even though LLVM has fewer years of development than gcc, it already gets better results in these two tests. In my opinion, the main LLVM strength is its modular and well designed architecture. It makes much easier to contribute to LLVM than to contribute to GCC. This shows how a well designed compiler architecture can improve the community software development speed.

Another interesting thing to be investigated is, at least in this two test cases, the link-time optimizations don’t improve the performance. Maybe in other test cases, we can get better link-time optimizations results.

These results show that we can expect a lot from the LLVM future.