On Monday 10 January 2005 2:53 pm, Olav Kongas wrote: > > I am also preparing for running usbtest on the driver.
I'm glad to see both of you looking at this ... "usbtest" is a good baseline. The other thing I'd recommend is using the "usbnet" driver with some network device, and doing TTCP stress tests; that involves less predictable traffic loads, and bidirectional ones. > I didn't manage to find a dongle with ezusb chip, but I got a > pci card with net2280 usb device chip and soon I'll try to > get it running the "device end" of the test system. I > haven't had time to take a closer look at the usbtest suite > yet. Therefore right now I cannot comment on your test > results. Net2280 makes a nice test harness, since you have full access to Linux within the "peripheral". You can insert arbitrary errors to see how the host behaves, and trace the "other end" of the protocol stack easily too. One minor caution though: some updates from Netchip seem to have borked the ability to switch gadget drivers using "rmmod g_zero; modprobe g_ether" and so forth. (By getting rid of a reset that showed up in some USB-IF testing; the reset seems to have been important in other ways.) So you may need to reboot after switching drivers ... if you boot and just stick to one gadget driver (like products based on this driver do!) then you should be OK. That's only for the recent 2.6 kernels; 2.4 based "peripherals" should still be fine. - Dave ------------------------------------------------------- The SF.Net email is sponsored by: Beat the post-holiday blues Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek. It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt _______________________________________________ [email protected] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
