ati-remote strangeness from 2.6.12 onwards
A message that is not in the archive (sent off-list or not yet received): [email protected]
I discovered a minor change in 2.6.10-mm1, changing this value back corrects the "ok" button issue.
diff -urN linux/drivers/usb/input/ati_remote.c linux-2.6.11/drivers/usb/input/ati_remote.c
--- linux/drivers/usb/input/ati_remote.c 2005-08-02
17:56:26.000000000 +1200
+++ linux-2.6.11/drivers/usb/input/ati_remote.c 2005-08-02 17:54:34.000000000 +1200
@@ -263,7 +263,7 @@
{KIND_FILTERED, 0xe4, 0x1f, EV_KEY, KEY_RIGHT, 1}, /* right */
{KIND_FILTERED, 0xe7, 0x22, EV_KEY, KEY_DOWN, 1}, /* down */
{KIND_FILTERED, 0xdf, 0x1a, EV_KEY, KEY_UP, 1}, /* up */
- {KIND_FILTERED, 0xe3, 0x1e, EV_KEY, KEY_ENTER, 1}, /* "OK" */
+ {KIND_FILTERED, 0xe3, 0x1e, EV_KEY, KEY_OK, 1}, /* "OK" */
{KIND_FILTERED, 0xce, 0x09, EV_KEY, KEY_VOLUMEDOWN, 1}, /* VOL + */
{KIND_FILTERED, 0xcd, 0x08, EV_KEY, KEY_VOLUMEUP, 1}, /* VOL - */
{KIND_FILTERED, 0xcf, 0x0a, EV_KEY, KEY_MUTE, 1}, /* MUTE */Quoted 43 lines · Andrew Morton
Begin forwarded message:
Date: Sun, 31 Jul 2005 12:18:23 +1200
From: mdew <[email protected]>
To: linux-kernel <[email protected]>
Subject: ati-remote strangeness from 2.6.12 onwardsusing 2.6.11 everything works fine, Upgrading too 2.6.13-rc3 I noticed 2 errors,
(1) When setting the HZ rating too 250 or 100 will cause the driver to excessfully repeat keys/accelerate when pressing a button, making it unusable :(
(2) the "Ok" button no longer works in anything after and including 2.6.12-rc1 (I've tested upto 2.6.13-rc3), 2.6.11 works fine. xbindkeys doesnt register any "ok" key presses on 2.6.12-rc1 onwards.
2.6.11 xbindkeys responses (nothing shows up in -rc1)
mediabawx2:~# xbindkeys -mk Press combination of keys or/and click under the window. You can use one of the two lines after "NoCommand" in $HOME/.xbindkeysrc to bind a key.--- Press "q" to stop. ---
"NoCommand"
m:0x0 + c:36
Return
"NoCommand"
m:0x0 + c:36
ReturnThanks :)
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Hi!
Quoted 11 lines
I discovered a minor change in 2.6.10-mm1, changing this value back corrects the "ok" button issue.
diff -urN
linux/drivers/usb/input/ati_remote.clinux-2.6.11/drivers/usb/input/ati_remote.c
---linux/drivers/usb/input/ati_remote.c2005-08-02
17:56:26.000000000 +1200
+++linux-2.6.11/drivers/usb/input/ati_remote.c2005-08-02 17:54:34.000000000 +1200
Review line 269
Quoted 8 lines
(quoted diff) +1
-1@@ -263,7 +263,7 @@ {KIND_FILTERED, 0xe4, 0x1f, EV_KEY, KEY_RIGHT, 1}, /* right */ {KIND_FILTERED, 0xe7, 0x22, EV_KEY, KEY_DOWN, 1}, /* down */ {KIND_FILTERED, 0xdf, 0x1a, EV_KEY, KEY_UP, 1}, /* up */ - {KIND_FILTERED, 0xe3, 0x1e, EV_KEY, KEY_ENTER, 1}, /* "OK" */ + {KIND_FILTERED, 0xe3, 0x1e, EV_KEY, KEY_OK, 1}, /* "OK" */ {KIND_FILTERED, 0xce, 0x09, EV_KEY, KEY_VOLUMEDOWN, 1}, /* VOL + */ {KIND_FILTERED, 0xcd, 0x08, EV_KEY, KEY_VOLUMEUP, 1}, /* VOL - */ {KIND_FILTERED, 0xcf, 0x0a, EV_KEY, KEY_MUTE, 1}, /* MUTE */
I'd say that KEY_ENTER is perhaps more logical there? It is certainly more usefull than "OK" key.
Pavel
Signature · 1 line
teflon -- maybe it is a trademark, but it should not be.
Hi Pavel,
Further testing, I initated xmodmap -e "keycode $X = p" (where X was 10->255), so theoritcally, all the buttons on the ati-remote should be mapped to "p". I found the TV Button, The DVD Button, the CH-/+ and the OK Button all non-working, every other button produced the "p". xbindkeys -mk doesnt respond to those particular keys either.
these changes occured after 2.6.11, which have caused it not to respond correctly
- {KIND_FILTERED, 0xc8, 0x03, EV_KEY, KEY_PROG1, 1}, /* TV */
- {KIND_FILTERED, 0xc9, 0x04, EV_KEY, KEY_PROG2, 1}, /* DVD */
+ {KIND_FILTERED, 0xc8, 0x03, EV_KEY, KEY_TV, 1}, /* TV */
+ {KIND_FILTERED, 0xc9, 0x04, EV_KEY, KEY_DVD, 1}, /* DVD */
- {KIND_FILTERED, 0xe3, 0x1e, EV_KEY, KEY_ENTER, 1}, /* "OK" */
+ {KIND_FILTERED, 0xe3, 0x1e, EV_KEY, KEY_OK, 1}, /* "OK" */couldnt get the channelup/down to work correctly, even reversing the orginal patch doesnt help.
- {KIND_FILTERED, 0xd1, 0x0c, EV_KEY, KEY_CHANNELUP, 1}, /* CH + */
- {KIND_FILTERED, 0xd0, 0x0b, EV_KEY, KEY_CHANNELDOWN, 1},/* CH - */
+ {KIND_FILTERED, 0xd0, 0x0b, EV_KEY, KEY_CHANNELUP, 1}, /* CH + */
+ {KIND_FILTERED, 0xd1, 0x0c, EV_KEY, KEY_CHANNELDOWN, 1},/* CH - */
however this one change is fine.
- {KIND_FILTERED, 0xea, 0x25, EV_KEY, KEY_PLAYCD, 1}, /* ( >) */
+ {KIND_FILTERED, 0xea, 0x25, EV_KEY, KEY_PLAY, 1}, /* ( >) */
My orginal patch was incorrect, heres a revised patch, fixes the OK button and the TV/DVD button issue. Would be nice to get it included into 2.6.13 ;-)
Quoted 26 lines · Pavel Machek
Hi!
Quoted 11 lines
I discovered a minor change in 2.6.10-mm1, changing this value back corrects the "ok" button issue.
diff -urN
linux/drivers/usb/input/ati_remote.clinux-2.6.11/drivers/usb/input/ati_remote.c
---linux/drivers/usb/input/ati_remote.c2005-08-02
17:56:26.000000000 +1200
+++linux-2.6.11/drivers/usb/input/ati_remote.c2005-08-02 17:54:34.000000000 +1200Review line 269
Quoted 8 lines
(quoted diff) +1
-1@@ -263,7 +263,7 @@ {KIND_FILTERED, 0xe4, 0x1f, EV_KEY, KEY_RIGHT, 1}, /* right */ {KIND_FILTERED, 0xe7, 0x22, EV_KEY, KEY_DOWN, 1}, /* down */ {KIND_FILTERED, 0xdf, 0x1a, EV_KEY, KEY_UP, 1}, /* up */ - {KIND_FILTERED, 0xe3, 0x1e, EV_KEY, KEY_ENTER, 1}, /* "OK" */ + {KIND_FILTERED, 0xe3, 0x1e, EV_KEY, KEY_OK, 1}, /* "OK" */ {KIND_FILTERED, 0xce, 0x09, EV_KEY, KEY_VOLUMEDOWN, 1}, /* VOL + */ {KIND_FILTERED, 0xcd, 0x08, EV_KEY, KEY_VOLUMEUP, 1}, /* VOL - */ {KIND_FILTERED, 0xcf, 0x0a, EV_KEY, KEY_MUTE, 1}, /* MUTE */I'd say that
KEY_ENTERis perhaps more logical there? It is certainly more usefull than "OK" key.
Pavel
Quoted 3 lines
--
teflon -- maybe it is a trademark, but it should not be.
drivers/usb/input/ati_remote.c +3 -3
@@ -252,8 +252,8 @@ {KIND_FILTERED, 0xdd, 0x18, EV_KEY, KEY_KPENTER, 1}, /* "check" */ {KIND_FILTERED, 0xdb, 0x16, EV_KEY, KEY_MENU, 1}, /* "menu" */ {KIND_FILTERED, 0xc7, 0x02, EV_KEY, KEY_POWER, 1}, /* Power */ - {KIND_FILTERED, 0xc8, 0x03, EV_KEY, KEY_TV, 1}, /* TV */ - {KIND_FILTERED, 0xc9, 0x04, EV_KEY, KEY_DVD, 1}, /* DVD */ + {KIND_FILTERED, 0xc8, 0x03, EV_KEY, KEY_PROG1, 1}, /* TV */ + {KIND_FILTERED, 0xc9, 0x04, EV_KEY, KEY_PROG2, 1}, /* DVD */ {KIND_FILTERED, 0xca, 0x05, EV_KEY, KEY_WWW, 1}, /* WEB */ {KIND_FILTERED, 0xcb, 0x06, EV_KEY, KEY_BOOKMARKS, 1}, /* "book" */ {KIND_FILTERED, 0xcc, 0x07, EV_KEY, KEY_EDIT, 1}, /* "hand" */
@@ -263,7 +263,7 @@ {KIND_FILTERED, 0xe4, 0x1f, EV_KEY, KEY_RIGHT, 1}, /* right */ {KIND_FILTERED, 0xe7, 0x22, EV_KEY, KEY_DOWN, 1}, /* down */ {KIND_FILTERED, 0xdf, 0x1a, EV_KEY, KEY_UP, 1}, /* up */ - {KIND_FILTERED, 0xe3, 0x1e, EV_KEY, KEY_OK, 1}, /* "OK" */ + {KIND_FILTERED, 0xe3, 0x1e, EV_KEY, KEY_ENTER, 1}, /* "OK" */ {KIND_FILTERED, 0xce, 0x09, EV_KEY, KEY_VOLUMEDOWN, 1}, /* VOL + */ {KIND_FILTERED, 0xcd, 0x08, EV_KEY, KEY_VOLUMEUP, 1}, /* VOL - */ {KIND_FILTERED, 0xcf, 0x0a, EV_KEY, KEY_MUTE, 1}, /* MUTE */
Hi 'mdew' (you do have a name, do you?),
Quoted 2 lines · mdew
mapped to "p". I found the TV Button, The DVD Button, the CH-/+ and the OK Button all non-working, every other button produced the "p".
Could you please try 'showkey -s' from a console on all of those keys?
Pavel: I would think that 'more useful' is not really the same as 'correct'. If you find it useful to map this key to 'ENTER', so you should remap it in userspace. It should not be KEY_ENTER in the kernel for at least two reasons:
- The key is labled 'ok' (and not enter). I assume the code KEY_OK is made for exactly that kind of key and certain applications might look for exactly this code.
- You might want to differentiate between this key and the ENTER key of your keyboard, at least I do. If the kernel is sending the same code for both keys, this is not possible in userspace.
Frank
Quoted 5 lines
Quoted 2 lines · mdew
mapped to "p". I found the TV Button, The DVD Button, the CH-/+ and the OK Button all non-working, every other button produced the "p".
Could you please try 'showkey -s' from a console on all of those keys?
Without my patch, nothing shows up when pressing OK, TV and DVD, in showkey -s
# with my patch mediabawx2:~# showkey -s kb mode was XLATE
press any key (program terminates after 10s of last keypress)...0x1c 0x9c - OK0xe0 0x1f 0xe0 0x9f - TV0xe0 0x17 0xe0 0x97 - DVD
Quoted 15 lines
Pavel: I would think that 'more useful' is not really the same as 'correct'. If you find it useful to map this key to 'ENTER', so you should remap it in userspace. It should not be
KEY_ENTERin the kernel for at least two reasons:- The key is labled 'ok' (and not enter). I assume the code
KEY_OKis made for exactly that kind of key and certain applications might look for exactly this code.
- You might want to differentiate between this key and the ENTER key of your keyboard, at least I do. If the kernel is sending the same code for both keys, this is not possible in userspace.Frank
Hi!
Quoted 3 lines
- You might want to differentiate between this key and the ENTER key of your keyboard, at least I do. If the kernel is sending the same code for both keys, this is not possible in userspace.
No, I think that you can still diferentiate between them ... they come from different keyboard after all. See /dev/input/event*.
Pavel
Signature · 1 line
teflon -- maybe it is a trademark, but it should not be.
Quoted 12 lines · mdew
I discovered a minor change in 2.6.10-mm1, changing this value back
corrects the "ok" button issue.diff -urN
linux/drivers/usb/input/ati_remote.clinux-2.6.11/drivers/usb/input/ati_remote.c
---linux/drivers/usb/input/ati_remote.c2005-08-02
17:56:26.000000000 +1200
+++linux-2.6.11/drivers/usb/input/ati_remote.c2005-08-02 17:54:34.000000000 +1200
@@ -263,7 +263,7 @@
Quoted 8 lines
{KIND_FILTERED, 0xe4, 0x1f, EV_KEY, KEY_RIGHT, 1}, /* right */
{KIND_FILTERED, 0xe7, 0x22, EV_KEY, KEY_DOWN, 1}, /* down */
{KIND_FILTERED, 0xdf, 0x1a, EV_KEY, KEY_UP, 1}, /* up */
- {KIND_FILTERED, 0xe3, 0x1e, EV_KEY, KEY_ENTER, 1}, /* "OK" */
+ {KIND_FILTERED, 0xe3, 0x1e, EV_KEY, KEY_OK, 1}, /* "OK" */
{KIND_FILTERED, 0xce, 0x09, EV_KEY, KEY_VOLUMEDOWN, 1}, /* VOL + */
{KIND_FILTERED, 0xcd, 0x08, EV_KEY, KEY_VOLUMEUP, 1}, /* VOL - */
{KIND_FILTERED, 0xcf, 0x0a, EV_KEY, KEY_MUTE, 1}, /* MUTE */This appears to be already applied in 2.6.12-rc5.
Sorry Andrew, but the diff was incorrectly made, the updated patch, reverts the changes too TV, DVD and OK Button, from 2.6.12-rc1.
diff -ruN linux-2.6.12/drivers/usb/input/ati_remote.c linux-2.6.12-modified/drivers/usb/input/ati_remote.c
--- linux-2.6.12/drivers/usb/input/ati_remote.c 2005-06-18
07:48:29.000000000 +1200
+++ linux-2.6.12-modified/drivers/usb/input/ati_remote.c 2005-08-03
09:54:48.000000000 +1200
@@ -252,8 +252,8 @@
{KIND_FILTERED, 0xdd, 0x18, EV_KEY, KEY_KPENTER, 1}, /* "check" */
{KIND_FILTERED, 0xdb, 0x16, EV_KEY, KEY_MENU, 1}, /* "menu" */
{KIND_FILTERED, 0xc7, 0x02, EV_KEY, KEY_POWER, 1}, /* Power */
- {KIND_FILTERED, 0xc8, 0x03, EV_KEY, KEY_TV, 1}, /* TV */
- {KIND_FILTERED, 0xc9, 0x04, EV_KEY, KEY_DVD, 1}, /* DVD */
+ {KIND_FILTERED, 0xc8, 0x03, EV_KEY, KEY_PROG1, 1}, /* TV */
+ {KIND_FILTERED, 0xc9, 0x04, EV_KEY, KEY_PROG2, 1}, /* DVD */
{KIND_FILTERED, 0xca, 0x05, EV_KEY, KEY_WWW, 1}, /* WEB */
{KIND_FILTERED, 0xcb, 0x06, EV_KEY, KEY_BOOKMARKS, 1}, /* "book" */
{KIND_FILTERED, 0xcc, 0x07, EV_KEY, KEY_EDIT, 1}, /* "hand" */
@@ -263,7 +263,7 @@
{KIND_FILTERED, 0xe4, 0x1f, EV_KEY, KEY_RIGHT, 1}, /* right */
{KIND_FILTERED, 0xe7, 0x22, EV_KEY, KEY_DOWN, 1}, /* down */
{KIND_FILTERED, 0xdf, 0x1a, EV_KEY, KEY_UP, 1}, /* up */
- {KIND_FILTERED, 0xe3, 0x1e, EV_KEY, KEY_OK, 1}, /* "OK" */
+ {KIND_FILTERED, 0xe3, 0x1e, EV_KEY, KEY_ENTER, 1}, /* "OK" */ {KIND_FILTERED, 0xce, 0x09, EV_KEY, KEY_VOLUMEDOWN, 1}, /* VOL + */
{KIND_FILTERED, 0xcd, 0x08, EV_KEY, KEY_VOLUMEUP, 1}, /* VOL - */
{KIND_FILTERED, 0xcf, 0x0a, EV_KEY, KEY_MUTE, 1}, /* MUTE */
Quoted 24 lines · Andrew Morton
Quoted 12 lines · mdew
I discovered a minor change in 2.6.10-mm1, changing this value back
corrects the "ok" button issue.diff -urN
linux/drivers/usb/input/ati_remote.clinux-2.6.11/drivers/usb/input/ati_remote.c
---linux/drivers/usb/input/ati_remote.c2005-08-02
17:56:26.000000000 +1200
+++linux-2.6.11/drivers/usb/input/ati_remote.c2005-08-02 17:54:34.000000000 +1200
@@ -263,7 +263,7 @@Quoted 8 lines
{KIND_FILTERED, 0xe4, 0x1f, EV_KEY, KEY_RIGHT, 1}, /* right */ {KIND_FILTERED, 0xe7, 0x22, EV_KEY, KEY_DOWN, 1}, /* down */ {KIND_FILTERED, 0xdf, 0x1a, EV_KEY, KEY_UP, 1}, /* up */ - {KIND_FILTERED, 0xe3, 0x1e, EV_KEY, KEY_ENTER, 1}, /* "OK" */ + {KIND_FILTERED, 0xe3, 0x1e, EV_KEY, KEY_OK, 1}, /* "OK" */ {KIND_FILTERED, 0xce, 0x09, EV_KEY, KEY_VOLUMEDOWN, 1}, /* VOL + */ {KIND_FILTERED, 0xcd, 0x08, EV_KEY, KEY_VOLUMEUP, 1}, /* VOL - */ {KIND_FILTERED, 0xcf, 0x0a, EV_KEY, KEY_MUTE, 1}, /* MUTE */This appears to be already applied in 2.6.12-rc5.
Quoted 3 lines · Ryan Brown
Sorry Andrew, but the diff was incorrectly made, the updated patch, reverts the changes too TV, DVD and OK Button, from 2.6.12-rc1.
hm, this was about 20 patches and 300 emails ago and I've forgotten what we're discussing. For poor old scatterbrains it really helps if people can maintain a description of their patch alongside the patch itself.
Quoted 18 lines
diff -ruN
linux-2.6.12/drivers/usb/input/ati_remote.clinux-2.6.12-modified/drivers/usb/input/ati_remote.c
---linux-2.6.12/drivers/usb/input/ati_remote.c2005-06-18
07:48:29.000000000 +1200
+++linux-2.6.12-modified/drivers/usb/input/ati_remote.c2005-08-03
09:54:48.000000000 +1200(quoted diff) +2
-2@@ -252,8 +252,8 @@ {KIND_FILTERED, 0xdd, 0x18, EV_KEY, KEY_KPENTER, 1}, /* "check" */ {KIND_FILTERED, 0xdb, 0x16, EV_KEY, KEY_MENU, 1}, /* "menu" */ {KIND_FILTERED, 0xc7, 0x02, EV_KEY, KEY_POWER, 1}, /* Power */ - {KIND_FILTERED, 0xc8, 0x03, EV_KEY, KEY_TV, 1}, /* TV */ - {KIND_FILTERED, 0xc9, 0x04, EV_KEY, KEY_DVD, 1}, /* DVD */ + {KIND_FILTERED, 0xc8, 0x03, EV_KEY, KEY_PROG1, 1}, /* TV */ + {KIND_FILTERED, 0xc9, 0x04, EV_KEY, KEY_PROG2, 1}, /* DVD */ {KIND_FILTERED, 0xca, 0x05, EV_KEY, KEY_WWW, 1}, /* WEB */ {KIND_FILTERED, 0xcb, 0x06, EV_KEY, KEY_BOOKMARKS, 1}, /* "book" */ {KIND_FILTERED, 0xcc, 0x07, EV_KEY, KEY_EDIT, 1}, /* "hand" */
Review line 269
Quoted 8 lines
(quoted diff) +1
-1@@ -263,7 +263,7 @@ {KIND_FILTERED, 0xe4, 0x1f, EV_KEY, KEY_RIGHT, 1}, /* right */ {KIND_FILTERED, 0xe7, 0x22, EV_KEY, KEY_DOWN, 1}, /* down */ {KIND_FILTERED, 0xdf, 0x1a, EV_KEY, KEY_UP, 1}, /* up */ - {KIND_FILTERED, 0xe3, 0x1e, EV_KEY, KEY_OK, 1}, /* "OK" */ + {KIND_FILTERED, 0xe3, 0x1e, EV_KEY, KEY_ENTER, 1}, /* "OK" */ {KIND_FILTERED, 0xce, 0x09, EV_KEY, KEY_VOLUMEDOWN, 1}, /* VOL + */ {KIND_FILTERED, 0xcd, 0x08, EV_KEY, KEY_VOLUMEUP, 1}, /* VOL - */ {KIND_FILTERED, 0xcf, 0x0a, EV_KEY, KEY_MUTE, 1}, /* MUTE */
IOW: what does this (wordwrapped!) patch do?
Hi,
Quoted 1 line · Andrew Morton
IOW: what does this (wordwrapped!) patch do?
It changes the keycode the kernel is sending for three keys. For normal keyboards there is usually no argument to which keycode to send. An 'a' would send the keycoe for an 'a'. This however is a remote control. The keys are labled 'OK', 'TV' and 'DVD'. Therefore the kernel currently sends the keycodes KEY_OK, KEY_TV and KEY_DVD. The patch changes this to KEY_ENTER, KEY_PROG1 and KEY_PROG2.
I do not know about the motivation of this patch, as the kernel currently _does_ send keycodes, maybe just not the ones the some users might want. IMHO this is an issue of remapping the keycodes in userspace and I would like to leave the kernel-codes alone. However, I might not see the whole problem here because it is working fine for me.
Btw, Pavel:
Quoted 2 lines
No, I think that you can still diferentiate between them ... they come from different keyboard after all. See
/dev/input/event*.
How can I tell the consoles of linux which keyboard to use? So far they all use all keyboards (which is my usual keyboard mixed with the remote control keys). (Yes, I searched google extensivly and no, I do not have X on that machine.)
Frank