This is a guide to the
fbench C example project included in the EMAC OE SDK.
The project contains a floating point benchmark which tests the accuracy and speed of floating point operations on the target systems. The testing application utilizes floating point intensive ray tracing and Fast Fourier Transform algorithms to stress the processor. This project contains excerpts from the
fbench project by John Walker of Fourmilab. See John Walker's Floating Point Benchmarks project homepage for more information.
fbench project builds two executables:
fbench is a trigonometry intensive floating point benchmark. It is a complete optical design raytracing algorithm without the user interface.
ffbench is a Fast Fourier Transform benchmark. It loops through a fast Fourier transform of a square matrix of complex numbers, reverses the transform and then checks the results.
Opening, Building and Uploading the Project Files
For information on opening the project from within Eclipse, please see Importing the EMAC OE SDK Projects with Eclipse. Then, follow Using the EMAC OE SDK Projects with Eclipse for information on how to build, upload and execute the example.
Makefile can be used with the
make command from the commandline to build and upload the example. For information on this method, please see Using EMAC OE SDK Example Projects.
EMAC SDK 5.X
For information on opening the project from within QtCreator, please see QtCreator: Adding Source Files. Then, follow Getting Started With Qt Creator for information on how to build, upload and execute the example.
CMakefile.txt can be used with the
cmake command from the commandline to build and upload the example. For information on this method, please see Getting Started with the EMAC OE SDK.
Usage and Behavior
fbench project is intended for use on C implementations that define
int as 32 bits or longer and permit allocation and direct addressing of arrays larger than one megabyte.
fbench program is executed from the console. It takes a single optional parameter.
Where <itercount> specifies the number of iterations to be performed, with 1,000 being the default.
For archival purposes you'll want to use a value slightly higher than 1,000.
root@som9g20:/tmp# ./fbench 2000 Ready to begin John Walker's floating point accuracy and performance benchmark. 2000 iterations will be made. Measured run time in seconds should be divided by 2 to normalise for reporting results. For archival results, adjust iteration count so the benchmark runs about five minutes. Press return to begin benchmark:
fbench has finished it prompts the user to stop the timer (by pressing return).
Stop the timer:
No errors in results.
fbench reports that no errors were found in the floating point operations.
A Note on Suspicious Systems
The default functionality as described above is for systems that can be trusted to be reliable. When working with a system that is suspected of having issues,
fbench can be compiled with
ACCURACY defined. This will generate a version that executes as an infinite loop, performs the ray trace and checks the results on every pass. All incorrect results will be reported. It will keep running until it is stopped manually (using, for instance, CTRL-C).
ffbench program is executed from the console. It takes no parameters.
root@som9g20:/tmp# ./ffbench 20 passes. No errors in results.
It runs until it is finished and reports what it discovered. In this case it performed 20 passes (the default, specified in code) and found no errors.
The time that it takes for this benchmark to run is an indicator of the performance of the board running it. When running it from a Bash shell, the execution time can be measured thusly:
fbench floating point benchmark C example tests the speed and accuracy of your floating point operations, and is interactive by default. The
ffbench example, on the other hand, is non-interactive by default and can be readily used both for benchmarking a board's floating point performance and to test the accuracy of its FPU.