problem of escaped scancodes

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi, Daniel P. Berrange,

Our user found that some keycode can not be handled well when
they try to map other keycode to xt_kbd keycode, when xt_kbd keycode
is an escaped scancode.

The xt_kbd keycode http://git.gnome.org/browse/gtk-vnc/plain/src/keymaps.csv
are come from x86_keycodes[] of linux/drivers/char/keyboard.c.
And x86_keycodes[] are:

static const unsigned short x86_keycodes[256] =
	{ 0,  1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15,
	 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31,
	 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47,
	 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63,
	 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79,
	 80, 81, 82, 83, 84,118, 86, 87, 88,115,120,119,121,112,123, 92,
	284,285,309,  0,312, 91,327,328,329,331,333,335,336,337,338,339,
	367,288,302,304,350, 89,334,326,267,126,268,269,125,347,348,349,
	360,261,262,263,268,376,100,101,321,316,373,286,289,102,351,355,
	103,104,105,275,287,279,258,106,274,107,294,364,358,363,362,361,
	291,108,381,281,290,272,292,305,280, 99,112,257,306,359,113,114,
	264,117,271,374,379,265,266, 93, 94, 95, 85,259,375,260, 90,116,
	377,109,111,277,278,282,283,295,296,297,299,300,301,293,303,307,
	308,310,313,314,315,317,318,319,320,357,322,323,324,325,276,330,
	332,340,365,342,343,344,345,346,356,270,341,368,369,370,371,372 };


There is no keycode in range [128, 256] in this table, which means
all escaped scancode "e0 ??" are encoded as "0x100 | ??" in this table.
Example, LeftCtrl is 0x1d, and RightCtrl is escaped: "e0 1d",
RightCtrl is encoded as "0x100 | 0x1d"(=0x11d=285) in this table.
The code also tell me the same information:

		code = x86_keycodes[keycode];
		if (!code)
			return -1;

		if (code & 0x100)
			put_queue(vc, 0xe0);
		put_queue(vc, (code & 0x7f) | up_flag);



But some other place, escaped scancode "e0 ??" are encoded as "0x80 | ??",
this encoding is more commonly used, and qemu uses this encoding,
RightCtrl is encoded as "0x80 | 0x1d"(=0x9d=157) in qemu:

    { 0x9d, "ctrl_r" },

        keycode = keycodes[i];
        if (keycode & 0x80)
            kbd_put_keycode(0xe0);
        kbd_put_keycode(keycode & 0x7f);


The problem:
keymaps.csv uses the first encoding while qemu uses the second one, it
causes "virsh send-key" command can't work when it sends escaped scancodes.

Example:
When do "virsh send-key domain KEY_RIGHTCTRL", qemu will receive keycode=285
which is not expected.

So I suggest to change keymaps.csv.
Covert the old
KEY_RIGHTCTRL,97,RightControl,0x3e,,,88,,285,228,VK_RCONTROL,0xa3,0x65,0x65
to new
KEY_RIGHTCTRL,97,RightControl,0x3e,,,88,,157,228,VK_RCONTROL,0xa3,0x65,0x65
(ditto for other escaped scancodes)

Or just add the new line to keymaps.csv followed by the old line, and make
285 and 157 can work at the same time.

Thought?

Thanks,
Lai.

--
libvir-list mailing list
libvir-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/libvir-list


[Index of Archives]     [Virt Tools]     [Libvirt Users]     [Lib OS Info]     [Fedora Users]     [Fedora Desktop]     [Fedora SELinux]     [Big List of Linux Books]     [Yosemite News]     [KDE Users]     [Fedora Tools]