On Sat, Nov 19, 2016 at 10:03 PM, Samuel Thibault < samuel.thibault@xxxxxxxxxxxx> wrote: > Hello, > > Just looking over the patch, it looks good. I don't think it needs to be > split in 4 pieces though, I'd say merge patches 1, 3, and 4 together, > where patch 3 would be modified to make dummy use spk_serial_out too, to > get a patch which does just one complete thing: make serial_out a > method. > > Then you'll have other patches introducing spk_serial_out_tty and making > some drivers use it. > > Also note that as Greg said, the patches you have just submitted do > not really make sense alone. I don't know if staging people feel like > applying them while it's only pavement for future work. If so, then > good; otherwise, then fine too: it's good that you have sent them so > that we could check how they look like, I just wanted to let you know > that they might not get applied yet, just because one may want to see > the future patches before applying what you have done so far. > Yeah makes sense. Perhaps should have asked around beforehand, as my experience for anything other than trivial patches is non-existent. > BTW, I have been looking at functions again, we'll also need > a spk_synth_immediate_tty that drivers can use instead of > spk_synth_immediate, in addition to the _in, and _out functions. For now > (i.e. to make the dummy driver work, and probably a few simple more), > the tty versions for _out and _synth_immediate will probably enough to > get something working and commitable to the main kernel. > Good point, will add tty version of spk_synth_immediate. > > Also, as mentioned, now the hard work is getting to open the tty from > the kernel :) That'd be in a spk_tty_synth_probe function that drivers > would use instead of spk_serial_synth_probe. > > Samuel > Thanks, Okash _______________________________________________ Speakup mailing list Speakup@xxxxxxxxxxxxxxxxx http://linux-speakup.org/cgi-bin/mailman/listinfo/speakup
