That'll be very dependant on the exact application, and compiler. If you just want a very rough data point I guess you could benchmark a sha512sum loop on a large file on tmpfs generated from /dev/urandom. It's as meaningless a number as most, but a lot less meaningless than BogoMIPS - all that measures is how fast the CPU can idle, which in most cases nowdays is going to be the same as the CPU clock speed.
Robert Moskowitz rgm@htt-consult.com wrote:
So now I have some approximation of F20 running on my cubieboard2. So I want to do a performance comparison, and simple BOGOmips from /proc/cpuinfo probably does not tell the whole story of RISC vs CISC.
So is there a tool available for F20 on both that I can use?
Want to see what I have here.
arm mailing list arm@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/arm
On Fri, 30 May 2014, Gordan Bobic wrote:
That'll be very dependant on the exact application, and compiler. If you just want a very rough data point I guess you could benchmark a sha512sum loop on a large file on tmpfs generated from /dev/urandom. It's as meaningless a number as most, but a lot less meaningless than BogoMIPS - all that measures is how fast the CPU can idle, which in most cases nowdays is going to be the same as the CPU clock speed.
Actually the bogomips is now calibrated with a timer, regardless of the CPU clock speed. The printing of the bogomips value has therefore been removed since it is even more meaningless than it used to be.
See https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=9...
Nicolas
-
Gordan Bobic -
Nicolas Pitre