Hi,
there are multiple bugs in current (2.5.56) hiddev.c:
1. hiddev_ioctl(): in the big switch(), several cases simply fall through
to the next case. Obviously, the final return 0 is missing.
2. HIDIOCAPPLICATION: hid->collection[i].usage might be negative. This
causes the ioctl() ti retur
Hello LKML, USB-ML, Vojtech!
The initialization order of hid.o and hiddev.o is wrong:
drivers/usb/hiddev.c:573 hiddev_connect() is called by
drivers/usb/hid-core.c:1225 hid_probe() before
drivers/usb/hiddev.c:665 hiddev_init() is called.
This results in hiddev_devfs_handle being NULL for each h
