>It probably wouldn't be too hard to write a driver that presented a >/dev/dsp* interface to the world and was attached to jack at the other >end. You could then run two and attach rtcmix or whatever to them. > >On the other hand it might encourage poeple to put off supporting jack >nativly and that would be a Bad Thing(TM).
there's no reason to do this. the ALSA "share" PCM type can support such things and it already exists. OSS-API using apps need to use LD_PRELOAD with libaoss to get access to it, however. we don't want to encourage anyone to continue using the OSS-API, if only for the very simple reason that its design (direct kernel calls; ok, direct libc calls that make kernel calls) makes it impossible to do this kind of "re-routing" without LD_PRELOAD. the kernel design has become much clearer since OSS started, and most kernel folk these days would seriously balk at putting conversion/mixing/mux-ing code into the kernel, even as a module. this rules out an OSS-like API (i.e using read/write/open/close/ioctl) because such API's require code that like to live in the drivers. --p _______________________________________________ Alsa-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/alsa-devel
