Am Wed, 27 Apr 2011 15:19:16 -0400 schrieb Jarod Wilson <jarod@xxxxxxxxxxxx>: > Heh, we'll get there... I hope so. > I meant dmesg output after pressing the button that results in the > ir-keytable 4eb02 output... If I had to guess though, that was from > the "1" key on your remote, and the issue here we're facing is that > the keymap only has the last part of the scancode for lookup... It indeed was the "1" key. But the last messages in dmesg are: [26479.026437] ir_nec_decode: NEC decode failed at state 0 (17500us space) [26479.026447] ir_rc5_decode: RC5(x) decode failed at state 1 (17500us space) [26479.026454] ir_rc6_decode: RC6 decode failed at state 0 (17500us space) [26479.026460] ir_jvc_decode: JVC decode failed at state 0 (17500us space) [26479.026466] ir_sony_decode: Sony decode failed at state 0 (17500us space) I don't think they are related to the keystrokes. > However, I think I do at least see why you have no active protocols. > It looks like the v4l-utils ir-keytable rule is loading a new map > (probably the terratec_cinergy_xs one), which doesn't have a specific > protocol listed, so no protocols get enabled. Judging by this feature request there doesn't seems to be any active protocol for every device, not only for the cx88: https://bugs.archlinux.org/task/23673 > Mauro, what's the expected behavior of ir-keytable when it loads a > keymap that says "type: UNKNOWN"? > > Heiko, here's something to try: > > Make a backup copy of /etc/rc_keymaps/terratec_cinergy_xs, and then > alter the original, so that it says "type: NEC" and prefix each of > the scancodes with 4eb (i.e., 0x41 KEY_HOME -> 0x4eb41 KEY_HOME). > Then load that, and see if things start actually working... (I'm > sort of shooting in the dark here, but I think its worth a try). It already said "type: NEC". But I ran `sed -i "s:x14:x4eb:g" /etc/rc_keymaps/nec_terratec_cinergy_xs` so that it says e.g. 0x4eb02 KEY_1 instead of 0x1402 KEY_1. And now it spits out a bit more, but I'm still getting scancodes only very randomly. When pressing the "1" key, ir-keytable -t now gives me: Testing events. Please, press CTRL-C to abort. 1303935368.238345: event MSC: scancode = 4eb02 1303935368.238373: event key down: KEY_1 (0x0002) 1303935368.238376: event sync 11303935368.278350: event MSC: scancode = 4eb02 1303935368.294324: event MSC: scancode = 4eb02 1303935368.390373: event MSC: scancode = 4eb02 1303935368.398335: event MSC: scancode = 4eb02 1303935368.648245: event key up: KEY_1 (0x0002) 1303935368.648258: event sync But, like I said before, it doesn't react always. Let's say, if I press the keys about 10 times I only get 2 or 3 scancodes, if not less. And dmesg now says this: [27163.270751] ir_nec_decode: NEC (Ext) scancode 0x04eb02 [27163.270759] rc_g_keycode_from_table: cx88 IR (TerraTec Cinergy 1400 : scancode 0x4eb02 keycode 0x02 [27163.270785] ir_rc5_decode: RC5(x) decode failed at state 1 (11500us space) [27163.270791] ir_rc6_decode: RC6 decode failed at state 0 (11500us space) [27163.270797] ir_jvc_decode: JVC decode failed at state 0 (11500us space) [27163.270802] ir_sony_decode: Sony decode failed at state 0 (11500us space) [27163.520656] ir_do_keyup: keyup key 0x0002 Heiko -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
