GCC Link-Time Optimizations On The ARM Cortex-A9

Written by Michael Larabel in Arm on 2 March 2013 at 02:07 PM EST. 1 Comment
ARM
The ARM Linux benchmarks continue. This time around we're looking at how the GCC LTO (Link-Time Optimization) performance fairs when running from a quad-core NVIDIA Tegra 3 Cortex-A9 tablet.

Previously I've looked closely at the GCC Link-Time Optimization Performance for x86 hardware and last month for LTO improvements in GCC 4.8. In this posting are some GCC LTO benchmarks being put out for reference from a NVIDIA Tegra 3, quad-core Cortex-A9 1.4GHz SoC.

This is just the most recent ARM Linux benchmarks on Phoronix. For the week already have been:

Benchmarking Ubuntu Linux On The Google Nexus 10 - This is the big article looking at the performance of the Google Nexus 10 (Samsung Exynos 5 Dual; ARM Cortex-A15 SoC) running Ubuntu Touch compared to numerous other x86 and ARM devices. Benchmarking of the Google Nexus 7 from Ubuntu Touch is forthcoming.

ARM Cortex-A15 Compiler Optimizations - Looking at various GCC compiler optimizations (sans LTO) from the new ARM Cortex-A15.

GCC 4.8 To Bump Up Speed On NVIDIA's Tegra 3 - GCC 4.8 will offer up some speed improvements for this quad-core A9 SoC.

GCC 4.8 Compiler Performance On ARM Cortex-A15 - Seeing how well GNU Compiler Collection 4.8 compares to existing GCC releases for the Cortex-A15 from the Samsung Exynos 5 Dual.

Plus other ARM Linux articles. In the coming days will be the talked about Google Nexus 7 results from running Ubuntu Touch, plus numerous other Linux benchmarks still being planned. At the moment it's still undecided whether to benchmark the Google Nexus 4 with its Qualcomm "Krait" Snapdragon S4 Pro SoC, if you're interested in those numbers, read this information. You can also see the Phoronix February stats for all the work that's involved.

Anyhow, the ARM GCC LTO benchmarks for your Saturday viewing can be found in full on OpenBenchmarking.org within 1303016-PTS-GCC48ARM14. Embedded below are the highlights from this benchmarking from the NVIDIA Tegra 3 Cardhu tablet.
ARMv7 Cortex A9 GCC 4.7 LTO Link-Time Optimizations
ARMv7 Cortex A9 GCC 4.7 LTO Link-Time Optimizations
ARMv7 Cortex A9 GCC 4.7 LTO Link-Time Optimizations
There are some performance improvements to see out of GCC Link-Time Optimizations from GCC 4.7.2. This feature is enabled via the "-flto" compiler switch that can be set via the CFLAGS/CXXFLAGS.
ARMv7 Cortex A9 GCC 4.7 LTO Link-Time Optimizations
ARMv7 Cortex A9 GCC 4.7 LTO Link-Time Optimizations
The compile time obviously increases when applying link-time optimizations.
ARMv7 Cortex A9 GCC 4.7 LTO Link-Time Optimizations
Continue on to the rest at OpenBenchmarking.org, a component of the open-source Phoronix Test Suite software.
Related News
About The Author
Michael Larabel

Michael Larabel is the principal author of Phoronix.com and founded the site in 2004 with a focus on enriching the Linux hardware experience. Michael has written more than 20,000 articles covering the state of Linux hardware support, Linux performance, graphics drivers, and other topics. Michael is also the lead developer of the Phoronix Test Suite, Phoromatic, and OpenBenchmarking.org automated benchmarking software. He can be followed via Twitter, LinkedIn, or contacted via MichaelLarabel.com.

Popular News This Week