> how can i transform ... *snip* ... > which takes about 3000 cycles into something FASTER. > I mean HOW can i efficiently write (from kernel level) into "info" > buffer whithout that much cycles spent??? Those are two different questions. Which do you want to do, SOMETHING faster, or the SAME THING faster? What about saving the raw timing information while the test is running, then converting it into a readable format after the test is complete? Do you have any hardware outputs available which could be controlled with only a few instructions? (CPU pins? Parallel port control or data lines? Serial port flow control lines? Ethernet port whose link you can enable and disable? PC speaker?) You could assert the output line when your packet enters the test zone, deassert it line when the packet leaves, and measure the interval with an oscilloscope. (Or maybe your test interval is too short for that...) - : send the line "unsubscribe linux-net" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
