On Mon, Apr 27, 2009 at 11:01:53AM +0800, Xiaofan Chen wrote: > On Mon, Apr 27, 2009 at 10:51 AM, Xiaofan Chen <xiaofanc@xxxxxxxxx> wrote: > > On Mon, Apr 27, 2009 at 10:39 AM, Kay Sievers <kay.sievers@xxxxxxxx> wrote: > >>>> Yeah, sounds fine. Or maybe wait until we dropped the dependency on > >>>> /proc? It still says: > >>>> ./lsusb -t > >>>> cannot open /proc/bus/usb/devices, No such file or directory (2) > >>>> > > > > There is a patch in Ubuntu. > > https://bugs.launchpad.net/debian/+source/usbutils/+bug/159189 > > But I think it is better just to remove it. Better yet, make the option work with out it at all, it is not that difficult to do. > >> I just looked at the new libusb and it seems it already parses the > >> "descriptors" file. If that works out, and is useful, the first thing > >> would be to convert to the new lib, I guess, and start from there. In thinking about it, it might be easier to not use lsusb if we use libudev for device enumeration. But I'll play around with it to see if it's easier or not... > > Great idea. ;-) Actually I proposed this earlier but I do not know much > > about programming. ;-( > > http://www.nabble.com/Re%3A--linux-usb-devel--libusb-silently-ignores-permission-errors-td23094681.html > > http://www.nabble.com/usbview-to-be-removed-from-Ubuntu-or-maybe-Debian--td23107766.html > > > > One more suggestion, currently lsusb dees not dump HID report > descriptor properly since it does not detach all the kernel HID > drivers. Many of the "lsusb -vvv" reports shows the same problem > with something like this "Report Descriptors: ** UNAVAILABLE **". > > I've the same problem last time. > http://www.nabble.com/How-to-dump-HID-report-descriptor-under-Linux-td19609562.html > Alan Stern suggests that we can manually unbind the HID interface > and then the report descriptor can be properly dumped. Maybe > the new usbutils can be updated to support this by > detach and then re-attach the kernel driver. That wouldn't be good, do you really want your keyboard driver unbound and then bound back? It might get a bit messy very quickly... I wonder if we can just provide a "hid_descriptors" file for the HID driver to offer up all of the HID descriptors as well, to prevent this issue from happening... thanks, greg k-h -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
