Hello, On 13-Jul-01 Alan Cox wrote: >> But there is a second, more fundamental problem: I dont know WHICH driver >> claims the audio interface. It is theoretically possible that a second >> audio >> driver is written, with an entirely different structure. Since >> private_data >> is a void* pointer and there is no way to identify the driver attached to >> it, I cannot rely on anything, _even_ a function exported by the current >> audio driver. > > Right - you'd want to talk to audio and have a direct function call > exported > by it that you can use to ask it. Then you can set up the data and provide > struct video_unit data via VIDIOCGUINT reporting the device minors, and > whether you have GAUDIO do meaningful stuff is also optional VIDIOCGUNIT? Hmm, when did you add that call to the document at http://roadrunner.swansea.uk.linux.org/v4lapi.shtml? 5 minutes ago? :))) [It´s in some bttv drivers already, so it has been around for a while longer; but I don´t recall seeing it... But it is exactly the ioctl call I need]. Now only a way to get that darn minor number... :) The problem with GAUDIO is, that there is a way to get the number of audio channels (usually 1). So a program would be tempted to call VIDIOCGAUDIO with a .audio parameter ranging from 0 to max_audio_channel - 1 (and don´t pay attention to what´s being returned). - Nemosft ----------------------------------------------------------------------------- Try SorceryNet! One of the best IRC-networks around! irc.sorcery.net:9000 URL: never IRC: nemosoft IscaBBS (bbs.isca.uiowa.edu): Nemosoft >> Never mind the daylight <<