On Sun, May 09, 2010 at 01:23:36PM +0200, Alexandre Ratchov wrote: > On Thu, May 06, 2010 at 11:44:37PM +0000, Jacob Meuser wrote: > > > > This one is very fun to test with > > aucat -l -m play,mon > > and two instances of gpsk31, one talking to the other. > Possibly playing music at the same time to add ``background > noise''. > > Using non-blocking mode for recording eats ~100% CPU, so > I added the below hack. > > [...] > > there are a lot of very easy but very intrusive changes that > could simplifty and overally improve gpsk31. > > Who is going to maintain these changes? Upstream doesn't > seem to exist anymore.
I banged my head on this one for a while because the call to the function to pass the audio fd to the transmitter was in the wrong place. it somehow worked for an int as a fd but did not work with the pointer to struct sio_hdl. *shrug* did you also see where it was only opening the device O_RDONLY or O_WRONLY but then trying to both read() and write() if the device claimed to be capable of full-duplex? -- jake...@sdf.lonestar.org SDF Public Access UNIX System - http://sdf.lonestar.org
