Re: Is irsend supported with devinput?
Alec Leamas <[email protected]>
| Newsgroups | gmane.comp.hardware.lirc |
|---|---|
| Message-ID | <[email protected]> |
Hi! On 20/06/17 11:23, Piotr Oniszczuk wrote: > This effectively excludes very effective setup where all IR subsystem > configuration for receiving and sending will be minimized to practically > zero effort. Yes. But the kernel stuff is out of control for us. > I had another idea: > -use devinput for receiving (lircd with devinput driver + lircd.conf > generated from linux.h) > -use default driver for sending > > Both lircd instances are using different sockets, protocols (RC6 and > LIRC) and inputs (inputX and lirc0). Yes. But they will still collide on the actual kernel/physical device which leads to problems... > any usage irsend turnoff RC6 protocol - so receiving stops working. > Manual enabling RC6 brings back receiving - and again - any usage > irsend turnoffs again RC6 protocol. ... like these > It looks like usage of default driver automatically turnoffs any other > protocols than LIRC. Is this behavior expected? More or less, yes. Accessing the same kernel device from two different drivers is, well, not strictly defined. This is more about the kernel than lirc. > If there will be way to disable this other-than-LIRC protocol turningoff > - then maybe solution I'm proposing will offer final goal (IR zero config) Realistically, there are really only one zero-config usecase supporting blasting: The default driver + an existing lircd.conf file available using irdb-get(1) or so. And hardware with rc (i. e., /dev/lirc0) kernel support). OTOH, most users don't do blasting. For these, the devinput driver supports anything supported by the kernel. This is IMHO a big enough userbase to care about. Cheers! --alec ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot