> does anybody get less than 10 sec for > > dd -if /dev/sdU0.0/data -of /dev/null -count 1024 -bs 65536 ? > > or even less than 20 sec?
10.4 sec on a sheevaplug (with 1 microframe interrupt threshold). By the way, everyone has been very tactful not to point out my silly mistake in saying: > ... You can use time(1) while > reading usb disk to see if cpu is really the bottleneck, or if > in fact your machine is spending most of its time waiting for > a usb interrupt. Of course most of the user+sys cpu time involved in reading from usb disk will be charged to the usb/disk process (normally a subprocess of usbd), not to user command doing the reading.
