On Mon, Feb 20, 2017 at 10:01 PM, Okash Khawaja <okash.khawaja@xxxxxxxxx> wrote: > > >> On 20 Feb 2017, at 21:52, Samuel Thibault <samuel.thibault@xxxxxxxxxxxx> wrote: >> >> Okash Khawaja, on lun. 20 févr. 2017 21:49:31 +0000, wrote: >>> Keeping spk_ttio_out() as it is also means that it mirrors >>> spk_serial_out behaviour more closely while having few side effects, >>> i.e. restarting of TTYs. >> >> Actually, spk_serial_out does have the side effect of restarting TTYs >> etc. on error, that's done by the spk_wait_for_xmitr() call. That's >> actually one of the reasons why I was thinking that spk_ttyio_out should >> be doing it. > > Ah okay. Right in that case it's actually better now because that side effect will at least be explicit. > > Thanks > >> >> Samuel Should spk_ttyio_out() also take in synth instance as its first argument? That will ofcourse mean changing signature of serial_out method inside spk_synth struct. Currently it uses a global synth instance which gets asigned in do_synth_init() which is fine but leaves one wondering where it came from. _______________________________________________ Speakup mailing list Speakup@xxxxxxxxxxxxxxxxx http://linux-speakup.org/cgi-bin/mailman/listinfo/speakup
