command reads its arguments two at a time, each pair of arguments
consisting of a scancode (given in hexadecimal) and a keycode (given
in decimal). For each such pair, it tells the kernel keyboard driver
to map the specified scancode to the specified keycode.
This command is useful only for people with slightly unusual keyboards,
that have a few keys which produce scancodes that the kernel does not
The usual PC keyboard produces a series of scancodes for each
key press and key release. (Scancodes are shown by
showkey -s, see showkey(1).)
The kernel parses this stream of scancodes, and converts it to
a stream of keycodes (key press/release events).
(Keycodes are shown by showkey.)
Apart from a few scancodes with special meaning, and apart from
the sequence produced by the Pause key, and apart from shiftstate
related scancodes, and apart from the key up/down bit,
the stream of scancodes consists of unescaped
scancodes xx (7 bits) and escaped scancodes e0 xx (8+7 bits).
It is hardwired in the current kernel that in the range 1-88
(0x01-0x58) keycode equals scancode. For the remaining scancodes
(0x59-0x7f) or scancode pairs (0xe0 0x00 - 0xe0 0x7f) a
corresponding keycode can be assigned (in the range 1-127).
For example, if you have a Macro key that produces e0 6f according
to showkey(1), the command
setkeycodes e06f 112
will assign the keycode 112 to it, and then loadkeys(1) can be used
to define the function of this key.
In 2.6 kernels key codes lie in the range 1-255, instead of 1-127.
(It might be best to confine oneself to the range 1-239.)
In 2.6 kernels raw mode, or scancode mode, is not very raw at all.
The code returned by showkey -s will change after use of setkeycodes.
A kernel bug. See also
The keycodes of X have nothing to do with those of Linux.
Unusual keys can be made visible under Linux, but not under X.