Am Montag, 3. Januar 2005 23:49 schrieb David Brownell: > The "main thing" is to implement the various methods correctly; > there's a test framework to help make sure you're interpreting > the API specification (kerneldoc, mostly in <linux/usb_gadget.h>) > correctly. The problem is that "correctly" is going to depend > on lots of hardware-specific details, and you won't find the > relevant details without testing. > > If you have specific questions, feel free to ask them; otherwise > I suspect that this is a case where the only way to learn is > by doing. In this case, by debugging the driver, and making > it work with all the drivers in the 2.6.10 kernel tree. Start > with Gadget Zero, a reference "known good" driver; and all the > "usbtest" cases (#9, #10, #14 for control transfers, #1..#8 > for bulk in/out, and #13 for bulk endpoint halts); then use > the other gadget drivers, each one of them tends to uncover > a different set of issues. Thanks for helping. My main reason for asking was not to oversee any docs, which might have made work much easier. But the existing docs are quiet OK. And debugging so far wend very well.
Greetings Thomas -- Wo kämen wir hin, wenn alle sagten, wo kämen wir hin, und niemand ginge, um einmal zu schauen, wohin man käme, wenn man ginge. «Kurt Marti» ------------------------------------------------------- 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 _______________________________________________ linux-usb-devel@lists.sourceforge.net To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
