JFFS3 & performance

Thomas Gleixner tglx at linutronix.de
Thu Dec 16 16:02:50 EST 2004


On Thu, 2004-12-16 at 21:51 +0100, Jörn Engel wrote:
> real    0m0.395s
> user    0m0.326s
> sys     0m0.067s
> 
> The algorithm doesn't run that fast, though.

Makes sense, as the modulo in your loop is a provided by the div
instruction on x86. I used this on a CPU where no modulo result was
available.

tglx






More information about the linux-mtd mailing list