On Dec 19, 2014 9:48 PM, "Keith Packard" <<a href="mailto:keithp@keithp.com">keithp@keithp.com</a>> wrote: > > Jason Ekstrand <<a href="mailto:jason@jlekstrand.net">jason@jlekstrand.net</a>> writes: > > > >> + if (err != -EINVAL && err != -ENOSYS) { > > > > I'm not terribly familiar with the ioctls here, but why are we ignoring > > EINVAL? The previous patch made it sound like ENOSYS was the "I don't > > support this" error and EINVAL was a genuine error. > > We're treating EINVAL and ENOSYS the same; an indication that the driver > doesn't actually care about this information. > > I looked at the kernel source, and it returns -ENOSYS if there isn't a > driver hook for receiving the dirty rects; perhaps there's some notion > that -EINVAL would be another possible return value? Mostly, I didn't > want to break a potentially working case, just to make it treat -ENOSYS > exactly as it treated -EINVAL. Seems perfectly reasonable assuming EINVAL isn't supposed to be fatal which it doesn't sound like it is. > > Apart from this my limited X knowledge says it looks perfectly > > reasonable. > > You stand at the beginning on the path to X knowledge. In time, perhaps > you will gain wisdom enough to review Peter's patches in the input > subsystem. I also hope to become that wise someday. I don't. I've seen what understanding the input subsystem has done to some people. > > Reviewed-by: Jason Ekstrand <<a href="mailto:jason.ekstrand@intel.com">jason.ekstrand@intel.com</a>> > > Thanks! > > -- > <a href="mailto:keith.packard@intel.com">keith.packard@intel.com</a>