On Mon, Oct 20, 2014 at 5:12 PM, David Härdeman <david@xxxxxxxxxxx> wrote: > On 2014-10-19 12:21, Tomas Melin wrote: >> >> Change default setting for enabled protocols. >> Instead of enabling all protocols, disable all except lirc during >> registration. >> Reduces overhead since all protocols not handled by default. >> Protocol to use will be enabled when keycode table is written by >> userspace. > > > I can see the appeal in this, but now you've disabled automatic decoding for > the protocol specified by the keymap for raw drivers? So this would also be > a change, right? Yes, you are right. Atleast it potentially still could change expected behaviour. > > I agree with Mauro that the "proper" long-term fix would be to teach the > LIRC userspace daemon to enable the lirc protocol as/when necessary, but > something similar to the patch below (but lirc + keymap protocol...if that's > possible to implement in a non-intrusive manner, I haven't checked TBH) > might be a good idea as an interim measure? > I think it looks feasible to implement that way. I'll look in to it and send a new patch series. Tomas > > >> >> Signed-off-by: Tomas Melin <tomas.melin@xxxxxx> >> --- >> drivers/media/rc/rc-ir-raw.c | 3 ++- >> 1 file changed, 2 insertions(+), 1 deletion(-) >> >> diff --git a/drivers/media/rc/rc-ir-raw.c b/drivers/media/rc/rc-ir-raw.c >> index a118539..63d23d0 100644 >> --- a/drivers/media/rc/rc-ir-raw.c >> +++ b/drivers/media/rc/rc-ir-raw.c >> @@ -256,7 +256,8 @@ int ir_raw_event_register(struct rc_dev *dev) >> return -ENOMEM; >> >> dev->raw->dev = dev; >> - dev->enabled_protocols = ~0; >> + /* by default, disable all but lirc*/ >> + dev->enabled_protocols = RC_BIT_LIRC; >> rc = kfifo_alloc(&dev->raw->kfifo, >> sizeof(struct ir_raw_event) * MAX_IR_EVENT_SIZE, >> GFP_KERNEL); -- 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
