On Wed, 2009-11-25 at 23:30 +0100, Krzysztof Halasa wrote: 

Sean Young [off-list ref] writes:

quoted

Absolutely. There are a number of use cases when you want access to the 
space-pulse (i.e. IR) information.

I think nobody proposes otherwise (except for devices which can't pass
this info).

I think we were taking about such devices.

I have no objection that devices that *do* decode the protocol, they
should be handled inside kernel.

But devices that send raw pulse/space data should be handled in lirc
that will feed the data back to the kernel via uinput.

Best regards,
Maxim Levitsky