| Message ID | 20090812223327.GA6827@hardeman.nu (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Thu, Aug 13, 2009 at 12:33:27AM +0200, David Härdeman wrote: > This patch adds an EV_IR bit to allow input drivers to let userspace > know that the hardware is an infrared remote control transceiver. > > No dev->irbit is defined yet but I plan to add that later (which would > be used to indicate whether the hardware supports RX, TX, etc). > > If the patch is accepted I'll send followup patches adding the > appropriate input_set_capability call to the relevant drivers. > Do you expect devices to actually _send_ EV_IR events?
On Thu, August 13, 2009 08:49, Dmitry Torokhov wrote: > On Thu, Aug 13, 2009 at 12:33:27AM +0200, David Härdeman wrote: >> This patch adds an EV_IR bit to allow input drivers to let userspace >> know that the hardware is an infrared remote control transceiver. >> >> No dev->irbit is defined yet but I plan to add that later (which would >> be used to indicate whether the hardware supports RX, TX, etc). >> >> If the patch is accepted I'll send followup patches adding the >> appropriate input_set_capability call to the relevant drivers. >> > > Do you expect devices to actually _send_ EV_IR events? Yes, and I have some patches going in that direction (loosely based on Jon Smirl's in-kernel IR driver patchset). The basic idea so far would be to add: IR_PROTOCOL IR_TOGGLE IR_DEVICE IR_SUBDEVICE IR_COMMAND Which should be sufficient at this point for the major IR protocols. Then input drivers can report to userspace the actual command they received (and hacks like using MSC_SCAN / MSC_RAW / the sysfs attribute you commented in my winbond-cir driver would not be necessary). It also allows userspace to know that a button was pressed even where there is no keymap for that particular remote which should help a lot in building custom keymaps. Further along we might also add IR_RAW to signal to willing user-space apps the raw IR timings for more esoteric remotes (I guess it would need to be enabled/disabled by an ioctl).
On Thu, Aug 13, 2009 at 10:32:16AM +0200, David Härdeman wrote: > On Thu, August 13, 2009 08:49, Dmitry Torokhov wrote: > > On Thu, Aug 13, 2009 at 12:33:27AM +0200, David Härdeman wrote: > >> This patch adds an EV_IR bit to allow input drivers to let userspace > >> know that the hardware is an infrared remote control transceiver. > >> > >> No dev->irbit is defined yet but I plan to add that later (which would > >> be used to indicate whether the hardware supports RX, TX, etc). > >> > >> If the patch is accepted I'll send followup patches adding the > >> appropriate input_set_capability call to the relevant drivers. > >> > > > > Do you expect devices to actually _send_ EV_IR events? > > Yes, and I have some patches going in that direction (loosely based on Jon > Smirl's in-kernel IR driver patchset). > > The basic idea so far would be to add: > > IR_PROTOCOL > IR_TOGGLE > IR_DEVICE > IR_SUBDEVICE > IR_COMMAND > > Which should be sufficient at this point for the major IR protocols. Then > input drivers can report to userspace the actual command they received > (and hacks like using MSC_SCAN / MSC_RAW / the sysfs attribute you > commented in my winbond-cir driver would not be necessary). > > It also allows userspace to know that a button was pressed even where > there is no keymap for that particular remote which should help a lot in > building custom keymaps. > > Further along we might also add IR_RAW to signal to willing user-space > apps the raw IR timings for more esoteric remotes (I guess it would need > to be enabled/disabled by an ioctl). > Would you mind posting the complete patchset?
Index: linux-2.6/drivers/input/input.c =================================================================== --- linux-2.6.orig/drivers/input/input.c 2009-08-12 19:45:10.000000000 +0200 +++ linux-2.6/drivers/input/input.c 2009-08-12 19:54:56.000000000 +0200 @@ -1375,6 +1375,10 @@ /* do nothing */ break; + case EV_IR: + /* do nothing */ + break; + default: printk(KERN_ERR "input_set_capability: unknown type %u (code %u)\n", Index: linux-2.6/include/linux/input.h =================================================================== --- linux-2.6.orig/include/linux/input.h 2009-08-12 19:56:03.000000000 +0200 +++ linux-2.6/include/linux/input.h 2009-08-12 19:57:53.000000000 +0200 @@ -98,6 +98,7 @@ #define EV_FF 0x15 #define EV_PWR 0x16 #define EV_FF_STATUS 0x17 +#define EV_IR 0x18 #define EV_MAX 0x1f #define EV_CNT (EV_MAX+1)
This patch adds an EV_IR bit to allow input drivers to let userspace know that the hardware is an infrared remote control transceiver. No dev->irbit is defined yet but I plan to add that later (which would be used to indicate whether the hardware supports RX, TX, etc). If the patch is accepted I'll send followup patches adding the appropriate input_set_capability call to the relevant drivers. Signed-off-by: David Härdeman <david@hardeman.nu> -- To unsubscribe from this list: send the line "unsubscribe linux-input" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html